Revisiting the Cambrian Burgess Shale Palaeocommunity in Light of New Field Discoveries from Marble Canyon, Kootenay National Park, British Columbia

Home , Acrothyra, Amiskwia, Amplectobelua, Burgess Shale, Canadaspis, Canadia (annelid)

Karma Nanglu

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Ecology and Evolutionary Biology University of Toronto

Karma Nanglu Doctor of Philosophy Department of Ecology and Evolutionary Biology University of Toronto 2019

ABSTRACT

The 508-million-year-old Burgess Shale (British Columbia) is among the most important fossil localities in the world as it provides a direct window into the Cambrian Explosion, the phenomenon whereby most metazoan groups appeared rapidly in the fossil record for the first time. For over 100 years, Burgess Shale fossils have provided unique insights into the early evolutionary history of animal life, but a holistic description of the community ecology of the

Burgess Shale has remained elusive. This dissertation aims to reinvestigate the Burgess Shale paleocommunity in light of the recently discovered Marble Canyon fossil site in Kootenay

National Park, integrating this new dataset with those from the type areas in Yoho National Park,

40km to the northwest (Walcott Quarry, Raymond Quarry, and Tulip Beds).

The first three chapters of this dissertation focus on the organismal level. An experimental decay study (Chapter 1) provides a framework for the interpretation of taphonomic bias in community reconstructions. The redescription of three enigmatic Cambrian taxa

(Chapters 2 and 3) provide novel phylogenetic and ecological information for understanding the patterns of diversity and niche structure at Marble Canyon. The community analysis (Chapter 4) represents the largest quantitative study of the Burgess Shale to date. Patterns of faunal stasis between most adjacent bedding assemblages followed by periodic variations in abundance and

ii species identity are found at both Marble Canyon and Walcott Quarry, the two best-studied sites with the finest level of stratigraphic/temporal data. Across the entire Burgess Shale paleocommunity, major shifts in representative taxonomic groups and ecological modes occur between major localities. The results of this work suggest that the Burgess Shale as a paleocommunity was highly heterogeneous in both its taxonomic and ecological composition.

Shifting abiotic and environmental variables are determined to be more significant in the long- term structuring of Burgess Shale communities than species interactions such as predation, which was previously considered paramount.

iii

ACKNOWLEDGEMENTS

The first thanks must go to my supervisor, Jean-Bernard Caron. To have unfettered access to what is, in my obviously unbiased opinion, one of the most significant and exciting fossil discoveries in the last half-century has been the opportunity of a lifetime, and I cannot over express my thanks to Jean-

Bernard for trusting me with this thesis. Moreover, to spend three months in the Burgess Shale and to be part of the process of paleontology from prospecting to discovery to analysis to seeing my work in the galleries of the Royal Ontario Museum has been an experience that is impossible to compare.

When I reflect on the core insight I’ve come away from this PhD with, the single word that comes to mind is scale. Unlike many of the other grad students in our department, being a paleontologist was not my childhood dream (forgiving, for a moment, the standard childhood shortsighted dalliance with dinosaurs), and so I was perhaps less prepared for thinking about deep time than most of my peers when I started. Working on the Cambrian explosion, then, is truly to be thrown in into the temporal deep end. I've come to realize that to work as a paleontologist is as close as you can come to being a time traveler, and I thank Jean-Bernard once again for giving me that chance.

I would like to acknowledge my committee members Dr. Hélène Cyr and Dr. Doug Currie. Their feedback and support for my research over the last 5 years has been an essential component of becoming the researcher I am now.

The Invertebrate Paleontology department at the Royal Ontario Museum has been a second home to me for the last 5 years. While the support of the entire museum has helped me get to where I am today,

I’ll quickly address each of various invertebrate workers in turn that I’ve had the pleasure of knowing:

Pete Fenton – Before I had even technically started as a graduate student, Jean-Bernard put me in contact with Pete as the be-all end-all problem fixer for my future studies. I put that reputation to the test in short order; I’ve since been told that our Paleobiology department rarely requires chambers for euthanizing, chilling, and rotting worms. This may explain our lack of a readily available incubator. Pete’s ability to

iv channel Macgyver left me with a faith in his problem solving abilities that would go unshaken to this day; the experiments for the first chapter of my thesis were facilitated by a bucket of water, an electric fan and a bag of rocks. Perhaps more importantly than that, Pete has always been an unfailingly supportive person to work with, and often a great sympathetic ear when academia has set you down the path of insanity.

Cédric Aria – I don’t think it would be a stretch to say that Cédric and I have had a disagreement or two in the past. But I’m certainly no stranger to arguments being my primary mode of communication and so in that way, I think we share a common language. I came to truly appreciate our conversations, and I can think of few people who put as much of themselves into their work as Cédric. There is a real sense of the auteur to Cédric, a wholehearted feeling that what he’s working on is of critical importance and that he’ll brook no compromise in what he thinks is the truth. I think that’s something that we can all admire, even if we don’t always agree. He’s also got a pretty good singing voice, and his mountainside yodeling is not to be despised.

Joe Moysiuk - I’m still sore that I didn’t make it into the acknowledgments for Joe’s first paper. But that’s only because I want to lay some claim to his future successes, which I’m sure will be manifold. The guy is pretty much designed to work on the Cambrian, it’s uncanny. He’s also a solid confidante during field work, which as anyone who’s spent time in the Burgess Shale will tell you, is absolutely crucial.

Sometimes, after a long day of back-breaking hiking, all you need is to be reminded of a burnt risotto to put your life back into perspective.

Maryam Akrami – I’m not sure that I love fossils half as much as Maryam. Sometimes I don’t know if anyone loves fossils half as much as Maryam. There have been more than a few moments where, feeling beaten down by grad school, I’ve been stunned by how much raw appreciation Maryam has for the material we’re working on. It really re-centers the importance of this type of work, which you can sometimes lose sight of among all the meetings and deadlines. I’d also say our world views differ pretty

v significantly, but I think Maryam is a person you can trust to speak her mind and stand by her convictions, which is something I really respect about her.

I would also like to thank Sebastian Kvist. I’ve learned a lot from Sebastian, and teaching the invertebrate diversity labs with him for the last three years has been a huge joy and privilege. Hanging out with him and his lab at the ROM has felt like spending time with extended family, and I can think of few people who are as unfailingly supportive. It’s really great to have a friend who you can trust will always be in your corner, and Sebastian is that kind of guy.

Grad school was the first time in my life where I felt like I was on the same wavelength as the people around me. I'd never had a group of peers who were as excited as I was for science, or wanted to spend their off-hours just reading about the most bizarre aspects of animal life. There are many to thank, but in alphabetical order I wanted to recognize: Erika Anderson, Sean Anderson, Victoria Arbour,

Viviana Astudillo, James Boyko, Rowshyra Castañeda, Kentaro Chiba, Hollis Dahn, Danielle Dufault,

Rafael Eiji Iwama, Michael Foisy, Bryan Gee, Scott Gilchrist, Jessica Hawthorn, Derek Larson, Aaron

LeBlanc, Mark MacDougall, Claire Manglicmot, Melanie Massey, Mateus Pepinelli, Ashley Reynolds,

Jade Simon, and Sarah Steele. Special thanks are also due to three fellow grad students in particular.

Melissa Orobko was a great support for me during many of the most difficult moments during my PhD. I recall one particularly long and unpleasant walk through the snow while I very seriously questioned whether or not I had it in me to stick it out in grad school. Melissa helped talk me down from that metaphorical ledge.

Thomas M. Cullen fits into a rare niche amongst the people I've met throughout grad school. That niche is being someone who I simultaneously respect as a peer, but also admire as an academic goal to strive towards. There are few people who I have been able to go to for such consistently good advice; you couldn’t ask for a better sounding board or friend. Perhaps even more importantly, few but Tom truly understand that:

“A true victory is to make your enemy see they were wrong to oppose you in the first place. To force them to acknowledge your greatness.”

Danielle de Carle is the exact kind of person I hoped I would meet in grad school. I’d also be hard pressed to come up with someone who makes for a better argument; I’m just not sure if it’s more fun to argue against her or with her. I can’t think of anyone else who makes me as excited about science, or really anyone who’s just as much fun to spend time with. How often do you meet someone who is equally as comfortable playing the Niles to your Frasier as they are the Clarice to your Hannibal?

Finally, the most significant thanks of all have to go to my whole family, but especially my parents. Their support for my scientific curiosity didn’t start with this PhD, it’s been present my entire life. It was there when my dad and I would watch documentaries together when I was a kid, or when he’d teach me about ants in our backyard. It was there in moments of doubt when my mom pushed me to put myself in new and uncomfortable situations to grow as a professional researcher. It’s given me the confidence to hike through the rainforests of Costa Rica, climb through the Rockies, and dive in the Indo-Pacific to satisfy my wonder for the natural world. All of my successes are owed to them.

vii

Table of Contents ABSTRACT...... ii

ACKNOWLEDGEMENTS...... iv

TABLE OF CONTENTS...... viii

LIST OF TABLES...... xii

LIST OF FIGURES...... xiii

Chapter 1...... xii

Chapter 2...... xvi

Chapter 3...... xviii

Chapter 4...... xxi

LIST OF APPENDICES...... xxvi

Chapter 2...... xxvi

Chapter 3...... xxx

Chapter 4...... xxxii

GENERAL INTRODUCTION...... xxxiiv

OVERVIEW OF CHAPTERS Chapter 1...... xlv Chapter 2...... xlvi Chapter 3...... xlvii Chapter 4...... xlviii

REFERENCES...... l

CHAPTER 1: Using experimental decay of modern forms to reconstruct the early evolution and morphology of fossil enteropneusts...... 1

Introduction...... 2

Materials and methods...... 5

Results...... 7

viii

Fossil comparisons...... 9 Proboscis...... 9 Nuchal skeleton and gill bars...... 10 Resistance of the esophageal organ...... 11 Outline and external features of dead specimens...... 11

Discussion...... 12 Phylogeny...... 12 Incongruities...... 15 Burgess Shale taphonomy...... 17

Conclusions...... 20

Acknowledgements...... 21

References...... 22

Tables...... 27

Figures...... 30

CHAPTER 2: Cambrian suspension-feeding tubicolous hemichordates...... 50

Background...... 51

Results...... 52

Discussion...... 55

Conclusions...... 57

Methods...... 58

Acknowledgements...... 59

References...... 60

Figures...... 66

Appendices...... 76

CHAPTER 3: A new Burgess Shale polychaete and the origin of the annelid head revisited...... 98

Summary...... 98

Results and discussion...... 100

Ecology...... 102

Phylogenetic analysis...... 103

The ‘‘Metameric Head’’ hypothesis...... 104

The ‘‘Chaetigerous Mouth’’ hypothesis...... 106

Acknowledgements...... 109

References...... 110

Figures...... 114

Appendices...... 122

CHAPTER 4: Diversity and structure of the Marble Canyon paleocommunity, Burgess Shale, British Columbia...... 150

Introduction...... 152

Materials and methods...... 154 The Marble Canyon locality, data collection and taphonomic considerations...... 154 Other localities...... 156 Diversity patterns and multivariate analyses...... 157 Ecological groups and functional diversity...... 160

Results...... 160 Taxonomic community diversity patterns at Marble Canyon...... 160 Patterns of ecological change at Marble Canyon...... 164 Burgess Shale multi-locality patterns...... 165

Discussion...... 168 The short time-scale structuring of the Burgess Shale community...... 168 Major patterns of community assembly across the Burgess Shale...... 170

Conclusions...... 175

Acknowledgements...... 176

References...... 177

Tables...... 184

Figures...... 186

Appendices...... 214

GENERAL CONCLUSIONS...... 252

List of Tables

Chapter 1

Table 1. Morphological characters of enteropneusts involved in the decay experiment (number in brackets indicates corresponding position in Fig. 2). Stages of decay are illustrated in Figures 4 and 5...... 27

Chapter 4

Table 1. Species indicator analysis for the Burgess Shale. The indicator statistic has two components. The A-component describes the likelihood of sampling from the indicated locality

(or group of localities) based on the presence of the indicator taxon. The B-component describes the likelihood of finding the indicator taxon when sampling from the indicated locality (or group of localities)...... 180

xii

List of Figures

Chapter 1

FIGURE 1. Phylogenetic tree of the hemichordates. The pterobranch Cephalodiscus has a single pair of gill pores, whereas gill slits with a collagenous skeleton are an enteropneust (and chordate) feature. Among the enteropneusts, genital wings and hepatic sacs are found in ptychoderids and some spengelids, but not harrimaniids. The nuchal skeleton is reduced or absent from the deep-sea torquaratorids. Red numbered bars indicate potential hypotheses for the phylogenetic position of Spartobranchus tenuis: 1, stem hemichordate; 2, stem pterobranch; 3, stem enteropneust; 4, stem harrimaniid (Caron et al. 2013)...... 30

FIGURE 2. A, Generalized anatomy of the Enteropneusta (lateral view) and sequence of decay of major features. The division of tripartite body plan (proboscis, collar, trunk) is highlighted.

Susceptibility of decay proceeds from dark red (gill pores and pre-oral ciliary organ) to light red, yellow, orange, light green, dark green (gill bars and nuchal skeleton). 1, Longitudinal proboscis muscles. 2, Proboscis coelomic cavity. 3, Heart-kidney-stomochord complex. 4, Pre-oral ciliary organ. 5, Mouth. 6, Nuchal skeleton. 7, Gill pores. 8, Gill bars. 9, Ventral midline. 10, Genital wings. 11, Esophageal organ. 12, Dorsal midline. 13, Hepatic sacs. 14, Digestive tube (genital wings and hepatic sacs are absent from Harrimaniidae). 15, Anus. B (i) Dorsal view; (ii) Dorsal view of the nuchal skeleton displaying characteristic “Y” shape (found internally and beneath the stomochord in the dashed-circled region); (iii) cross-section of the dashed-boxed region; gill bars curve laterally around the columnar trunk...... 32

xiii

FIGURE 3. Sequence and rate of decay for three enteropneust species. A, Saccoglossus pusillus.

B, Harrimania planktophilus. C, Balanoglossus occidentalis. Solid colors represent time intervals where the indicated level of decay was achieved for all trials. The tip of an arrow represents the first time interval where that level of decay was achieved. Green = level 1, yellow

= level 2, orange = level 3, red = level 4, white = level 5 (absent). Asterisk indicates earliest point of gill bar/nuchal skeleton disarticulation...... 34

FIGURE 4. Sequence of decay of Saccoglossus pusillus. Rows from top to bottom: proboscis, pharyngeal area, trunk. Columns from left to right: pristine, decay stages 1-4. Scale bars in rows

A and C, 1 cm, and in B, 1 mm. Time elapsed since euthanasia in hours in the top right of each cell...... 36

FIGURE 5. Sequence of decay of Harrimania planktophilus. Rows from top to bottom: proboscis, pharyngeal area, esophageal organ, trunk. Columns from left to right: decay stages 1-4. Scale bars, 1 mm. Time elapsed since euthanasia in hours in the top right of each cell...... 38

FIGURE 6. Proboscis morphology variations among S. tenuis specimens; scale bars, 1 mm. A, The various shapes that a proboscis may take after death that differ from the traditional “acorn” type morphology. B, A direct comparison among these shapes and variation in proboscis preservation.

C, Patchy local variation of muscle decay within the proboscis, compared with the patchy patterns of darkness/reflectivity in fossil taxa (D); A–C, clockwise from top left: ROMIP62124,

USNM 202841, ROMIP62123, USNM 202797; D, left to right: USNM 202183,

ROMIP63129...... 40

xiv

FIGURE 7. A, B, Preservation of clearly harrimaniid type nuchal skeletons (1) and gill bars (2). Scale bars

=1mm. Orientation of preservation may result in variable presentations of the nuchal skeleton. A: the nuchal skeleton is preserved on a plane parallel to the fossil face (ROM 63128). B: the nuchal skeleton is preserved at a slight angle to the fossil face (USNM 202469). C and D: Preservation of the esophageal organ. The presence of a dark/reflective oblong patch at the end of the pharyngeal trunk after the gill bars corresponds to the anatomical position of the esophageal organ. The esophageal organ is the most decay- resistant soft character. C, USNM 202472. D, USNM 509806...... 42

FIGURE 8. Tube structures in S. tenuis show high degrees of preservation with clearly defined edges, in contrast with the rapidly disintegrating mucus tubes of extant torquaratorids (clockwise from top left: ROMIP62129, ROMIP63126, USNM 202097, ROMIP63127). Scale bars, 1 cm.

Breakages terminate more abruptly than would be expected for a mucus-based structure...... 44

FIGURE 9. Incongruities between decay data and the fossil record (clockwise from top left:

ROMIP63124, USNM 202780, ROMIP63125). A, B, High degrees of pharyngeal decay relative to the proboscis and digestive trunk, an inversion of what would be expected. C, A nuchal skeleton structure, rotated 180 from its expected orientation. There is no evidence that the proboscis has been disturbed; thus the external factors that resulted in this orientation are unclear. D, The Mazon Creek taxon Mazoglossus ramsdelli; although the tripartite body outline of an enteropneust is unequivocal, there is no discernible preservation of internal features...... 46

xv lost during motion. 2, A comparatively well-preserved pharyngeal area displaying high gill bar clarity. 3, Significant post-pharyngeal damage to the trunk. 4, An unidentifiable, disarticulated tissue mass...... 48

Chapter 2

FIGURE 1. Schematic anatomy of Oesia disjuncta. Co: collar, Cr: circum-collar ridge, Dg: digestive groove, Pr: proboscis, Hks: heart-kidney-stomochord complex, Gb: gill bars, Gp: gill pores, Mo: mouth, Po: pores, Ps: posterior structure, Tr: trunk, Tu: tube. Dashed lines indicate transverse cross sections...... 66

FIGURE 2. General morphology of Oesia disjuncta from the Burgess Shale. (Specimens in d, e and jcome from the Walcott Quarry; all other specimens come from Marble

Canyon). a Note bilobed posterior structure and extended pharyngeal area (ROMIP63737, part and counterpart are superimposed at the dashed line). b, c Tripartite body plan and internal organs in the proboscis (ROMIP63711). d, e Large proboscis and possible nuchal skeleton

(USNM 509815), see also Additional file 5A–C. f Well-developed bilobed posterior structure

(ROMIP63713). g–i Details of the pharyngeal area (h, partial counterpart of g, highlighted by vertical dashed line; i is close-up of framed area in g, ROMIP63710). j Left and right pairs of gill bars preserved in lateral view (USNM 277844). Direct light images: a, b, h; polarized light images: c–g, j; SEM image: i. Co: collar, Cr: circum-collar ridge, Dg: digestive groove, Dm: dorsal midline, Gb: gill bars, Hks: heart-kidney-stomochord complex, Ll: lateral side left, Lr: lateral side right, Ns: nuchal skeleton, Pr: proboscis, Ps: posterior structure, Tr: trunk. Scale bars: a = 10 mm, b–e = 1 mm, f–h = 5 mm, i = 500 μm, j = 2 mm...... 68

xvi

FIGURE 3. Margaretia dorus tubes and associations with Oesia disjuncta from the Burgess Shale.

Specimens in a and d come from the Raymond Quarry; all other specimens come from Marble

Canyon. (a–h) Taphonomic gradient of the worm inside its tube from generally poorly preserved

(a) to better preserved (h); the tubes tend to preserve more poorly at Marble Canyon relative to tubes from the Raymond Quarry showing similar amounts of decay of the worm. a Holotype of M. dorus with worm preserved as a dark/reflective band along the central axis of the tube

(USNM 83922). b, c Small fragments of tubes containing worms showing only few recognizable features (b: ROM 63955, c: ROMIP63956). d Part of a tube excavated to reveal a poorly preserved worm inside (ROMIP63715). e Tripartite body plan recognizable but worm heavily decayed (ROMIP63953). f Clear posterior structure but indistinct proboscis and trunk

(ROMIP63957). g Poorly preserved trunk and faded tube (ROMIP63952). h Close-up of framed area in g on counterpart, showing gill bars readily visible. i, jSpecimen showing clear tripartite body plan and evidence of gill bars (ROM 63715). k The extant acorn worm Saccoglossus pusillus after 48 hours of decay at 25 °C showing dissociated parts, although the tripartite body plan is still recognizable. l, m O. disjunctaoutside of its tube, showing extreme signs of decay comparable with k. Direct light (l) is contrasted with polarized light (m) to reveal different aspects of fossil morphology (ROMIP63954). The ectoderm is fraying off, the proboscis is indistinct and the trunk has lost turgidity. Most worms preserved inside their tubes show a similar level of preservation. Direct light images: a, b, d, l; polarized light images: c, e–i, m. Bi: node of bifurcation, Fe: fibrous elements, Wo: worm, other acronyms see Figs. 1 and 2. Scale bars: a–c, e–g, k–m = 10 mm, d, i = 5 mm...... 70

xvii

FIGURE 4. a, b Spirally arranged pores perforate the tube (ROMIP63716; see also Additional file 9A). cTwo examples of multiple bifurcation points in a single specimen. Extreme size variation underscores the fragmentary nature of most tubes (left: KUMIP 204373, right: KUMIP

241392). d, e Tube showing three-dimensional preservation. d Large section of the tube has been broken off revealing the other side of the tube. e The broken segment has been placed back in its original configuration to illustrate the three-dimensionality of the tube (KUMIP

147911). f, g Close-up of the pores and fibrous texture of the tube. Individual fibres are micrometre small (ROM 63705). Bi: node of bifurcation, Fe: fibrous elements, Lo: lower surface, Po: pores, Up: upper surface, Wo: worm, other acronyms see Fig. 2. Scale bars: a, b, f, g = 5 mm, c–e = 10 mm...... 72

FIGURE 5.a Life reconstruction with hypothetical closed terminal ends of the tubes — part of one tube partially removed to show a worm (drawing by Marianne Collins). b Phylogenetic relationship of Deuterostomia derived from [2]. Mapping of characters based on [1, 2] with our proposed hypothetical position for Oesia disjuncta as a basal hemichordate (dashed line with question mark). The position of Spartobranchus tenuis is based on a taphonomic study of extant and fossil enteropneusts [11]. Character states: 1) pharyngeal gill bars, suspension feeding; 2) notochord; 3) tubicolous; 4) miniaturization, coloniality; 5) fuselli; 6) loss of tubicolous lifestyle, deposit feeding; 7) indirect development via tornaria larva; 8) stereom, water vascular system…………………………………………………………………………………………....74

Chapter 3

FIGURE 1. General morphology of Kootenayscolex barbarensis from the Burgess Shale (Marble

Canyon locality). All images oriented with anterior directed to the top of the page. A: holotype

xviii

(ROMIP64388), nearly complete specimen (posterior missing) showing well preserved palps, median antenna, and chaetae. B-C: paratype (ROMIP64389); B: anterior section showing well- preserved internal head features and sediment infill within the gut; C: elemental map showing phosphatized areas, interpreted as possible remnants of vascularized tissues, within the palps, head and basal portion of parapodia. A thin carbon line running from the front of the head to the palps represents putative neural tissues (see close up Figure 2A-B). D-E: paratype

(ROMIP64390); D: paratype alongside putative K. barbarensis juvenile specimens; E: detail showing parapodia preserved in darker black. F-G: paratype (ROMIP62972); F: specimen showing varying parapodial angles and proximal portions of chaetae; G: close up anterior section. H: two paratypes (ROMIP64391.1(left)-2(right)), two complete specimens with elongate chaetae, median antenna and palps. I-J: picture showing seven K. barbarensis specimens, including one paratype (ROMIP63099.1 - see star on figured specimen); J: close up showing a putative juvenile of K. barbarensis preserved with long chaetae. K-N: paratype (ROMIP64392.1- right); K: overall view of paratype with palps, median antenna and mud-filled gut extending through the entire body (paratype partially overlaying a second specimen to the left); L: close up of anterior gut section; M: close up of posterior gut section; N: close up of the left palp showing distal flattening. All images using cross-polarized light except C. Acronyms: an?- anus?, che- chaetae, det- degraded tissue, gin- gut infill, gut- gut, juv- juvenile, mot- mouth, nec- neurochaetae, nep- neuropodia, net- neural tissue, noc- notochaetae, nop- notopodia, mea- median antenna, pal- palp, par- parapodia, pco- prostomial coelom, pep- peristomial parapodia, pec- peristomial chaetae, ppe- proto-peristomium, pro- prostomium,. Scale bars = 1 mm. See also

Figures S1, S2, S3...... 114

xix

FIGURE 2. Head, parapodial and chaetal morphology of Kootenayscolex barbarensis from the

Burgess Shale (Marble Canyon locality). A-C: specimens showing putative location of mouth and internal tissues within palps and median antenna (C); A-B: close ups of boxed area in Fig.1B using direct light (A) and a composite line drawing of 1B and 1C (B); C: paratype

(ROMIP64393), superimposed images of both part and counterpart using Apply Image and overlay blending mode in Adobe Photoshop CS6. D-H: specimens showing palps, median antenna (except G) and peristomial chaetae directed anterio-laterally; D: close up of boxed area in Fig. 1A. E: line drawing of 2D. F: paratype (ROMIP64394). G: line drawing of 2F. H: paratype (ROMIP64395). I: close up of boxed area in Fig. 1H. J: close up of chaetal bundle. K:

ROMIP64396. L: close up of chaetal bundle. M: ROMIP64397. N: close up of chaetal bundle.

Cross-polarized light images (A-H, K, M); scanning electron microscope images (I, J, L, N). For acronyms see Fig. 1. Scale bars = 1 mm. See also Figures S1, S2, S3...... 116

FIGURE 3. Anatomy of Kootenayscolex barbarensis from the Burgess Shale. A: oblique view of the head. Dashed lines indicate cut-away transverse cross sections. Red: vascular tissue, blue: putative neural tissue, yellow: mouth and gut. For acronyms see Fig. 1. B: Life reconstruction.

Image © Royal Ontario Museum, Danielle Dufault. See also Figures S1, S2, S3...... 118

FIGURE 4. Evolutionary implications of Kootenayscolex barbarensis. A: Phylogenetic position of

K. barbarensis using majority rules Bayesian analysis (parameters are detailed in Star Methods).

Numbers at nods are posterior probabilities. Orange: taxa historically considered part of the major clade “Aciculata”; blue: taxa historically considered part of the ecomorphotype

“Sedentaria”; dark blue: Cambrian taxa; pink: sipunculids; green: mollusks. C-B: Two scenarios

xx for annelid head evolution which invoke modern developmental plasticity. B: modern polychaete head arising from a hypothesized ancestor with a biramous peristomium losing the notopodia and notochaetae to produce a uniramous Cambrian condition, then losing the first neuropodia and neurochaeta, leading to the crown group Annelida. C: alternatively, the uniramous peristomium of K. barbarensis and C. spinosa may have been produced during ontogeny (see text for descriptions of each numbered step). See also Figure S4...... 120

Chapter 4

FIGURE 1. Stratigraphic column of the Marble Canyon quarry and broad diversity trends. A: The top 10 taxa are a mix of different taxonomic groups, but are dominated by arthropods. The first four most abundant taxa are stratigraphically widespread throughout the quarry; the next 6 are more restricted (with the exception of Sidneyia). B: Major taxonomic groups represented at

Marble Canyon by abundance and diversity. Arthropods dominate both categories, but lophophorates, hemichordates, and annelids are significant components of the community.

Acronyms: ALGA – algae, ANNE – Annelida, ARTH – Arthropoda, CHORD – Chordata,

CTEN – Ctenophora, DINO – Dinocarridida, HEMI – Hemichordata, INDET – Indeterminate taxon, LOBO – Lobopodia, LOPH – Lophophorata, MOLL – Mollusca, OTHER – All other taxa not listed, PORI – Porifera, PRIA – Priapulida...... 182

FIGURE 2. Rarefaction curves for the four studied Burgess Shale localities at the level of bulk assemblages, and the individual Marble Canyon Quarry bedding assemblages. A: At the bulk assemblage level, rarefaction curves have generally plateaued, suggesting that they have been sampled sufficiently for diversity comparisons. B: Rarefaction curves for the Marble Canyon bedding assemblages, with extrapolated curves in dashed lines. For clarity, the 95% confidence

xxi intervals are only plotted for BA 350, BA 380, and BA 300. While the rank order of species richness changes slightly with extrapolated curves, the bounds of the confidence intervals make comparisons difficult to validate between any BAs except for the richest (BAs 340 and 350) and poorest (BAs 300 and 370)...... 184

FIGURE 3. Observed, inferred, and extrapolated species richnesses across BAs with 95% confidence intervals. Only BAs with greater than 299 specimens were included in quantitative analyses (BAs with fewer than 299 specimens are greyed out in the stratigraphic column to the right). Interpolating species richnesses to 299 specimens erases most patterns of fluctuating diversity. Extrapolating richness to 2,566 specimens suggests the same major trend as the observed pattern, but most bedding assemblages will fall within the confidence interval of the adjacent bedding assemblage...... 186

FIGURE 4. Rank abundance curves (also known as Whittaker plots) for the three species-richest and -poorest BAs. A lower slope, as seen in BAs 380, 400, and 350, indicates a greater evenness in abundance among constituent species...... 188

FIGURE 5. Proportions of major taxonomic groups among the Marble Canyon bedding assemblages. Lophophorates and hemichordates are the most abundant taxa in the upper levels of the quarry. After BA 340, arthropods become the most numerically dominant major taxon, which persists to the lowest strata of the quarry. Acronyms: ALGA – algae, ANNE – Annelida, ARTH

– Arthropoda, CHORD – Chordata, CTEN – Ctenophora, DINO – Dinocarridida, HEMI –

Hemichordata, INDET – Indeterminate taxon, LOBO – Lobopodia, LOPH – Lophophorata,

xxii

MOLL – Mollusca, OTHER – All other taxa not listed, PORI – Porifera, PRIA –

Priapulida...... 190

FIGURE 6. Taxonomic cluster analyses for the Marble Canyon Quarry using both the Morisita-

Horn and Jaccard indices. Both indices support the same broad clustering topology, with the

Upper Quarry (UQ) and Lower Quarry (LQ) BAs forming mutually exclusive clusters. The only significant difference is the shift in the location of BA 360 between the two major UQ clusters...... 192

FIGURE 7. Correspondence analysis for the Marble Canyon Quarry. Axis 1 delineates the lower quarry from the upper quarry and Axis 2 delineates the two sub-divisions within the upper quarry: the highest strata dominated by Oesia and Haplophrentis and the lower strata of the upper quarry dominated by Liangshanella and other arthropods...... 194

FIGURE 8. Proportions of major ecological groups among the Marble Canyon bedding assemblages. The upper levels of the quarry are largely composed of epibenthic-sessile- suspension feeders. As arthropods become more abundant, nektonic-vagrant-hunter/scavengers become the dominant niche. The lowest four BAs in the quarry are dominated by epibenthic- vagrant-deposit feeders. Acronyms: EPP – epibenthic primary producer, ESSU – epibenthic- sessile-suspension feeder, EVDE – epibenthic-vagrant-deposit feeder, EVHS – epibenthic vagrant hunter/scavenger, NKSU – nektobenthic vagrant suspension feeder, EVGR – epibenthic vagrant grazer, IVHS – infaunal vagrant hunter scavenger, NKDE – nektobenthic deposit feeder,

xxiii

NKHS – nektobenthic hunter scavenger, NKSU – nektobenthic suspension feeder, PEHS – pelagic hunter scavenger, PESU – pelagic suspension feeder...... 196

FIGURE 9. Cluster analysis for the Marble Canyon ecological abundance matrix using the

Morisita-Horn index. Three major clusters are recovered, corresponding to the uppermost levels of the Upper Quarry (UQ1), the lower levels of the Upper Quarry (UQ2), and the Lower Quarry

(LQ)...... 198

FIGURE 10. Ecological correspondence analysis for the Marble Canyon Quarry. There are three main types of ecological groups: those dominated by epibenthic-sessile-suspension feeding

(ESSU; purple) those dominated by nektobenthic-vagrant-hunter/scavengers (NKHS; blue) and those dominated nektobenthic-vagrant-deposit feeding (NKDE; pink)...... 200

FIGURE 11. Taxonomic cluster analyses for the Walcott Quarry (WQ), Raymond Quarry (RQ),

Tulip Beds (TB) and Marble Canyon (MC) using both the Morisita-Horn and Jaccard indices.

Localities cluster as reciprocally exclusive using the Jaccard index. The Morisita-Horn index recovers UQ2 BAs from MC as clustered with select WQ BAs, and the LQ BAs from MC as clustered with the TB locality...... 202

FIGURE 12. Detrended correspondence analysis for the Walcott Quarry (WQ: red), Raymond

Quarry (RQ: green), Tulip Beds (TB: purple) and Marble Canyon (MC: blue) taxonomic abundance matrix. Indicator species are plotted in the colour of their respective locality.

Indicator species for groups of more than one locality are plotted in black...... 204

xxiv

FIGURE 13. Ecological cluster analyses for the Walcott Quarry (WQ), Raymond Quarry (RQ),

Tulip Beds (TB) and Marble Canyon (MC) the Morisita-Horn index. Four main clusters are recovered dominated by three different trophic modes: suspension feeding (ESSU: green); hunting/scavenging (HS: blue); deposit feeding (DE: purple and NKDE: orange)...... 206

FIGURE 14. Ecological correspondence analysis for the Walcott Quarry (WQ), Raymond Quarry

(RQ), Tulip Beds (TB) and Marble Canyon (MC). Convex hulls are plotted and coloured according to the results of the ecological cluster analysis (Figure 11). For clarity, only highly recurrent ecological modes are plotted. See Figure 8 for acronyms...... 208

xxv

List of Appendices

Chapter 2

Appendix 1. High-resolution imagery of the original three Oesia disjuncta specimens figured by

C.D. Walcott, 1911, from the Burgess Shale (Walcott Quarry). (A–C) Lectotype (A, part; B, C, counterpart). The serial striations throughout the trunk initially lead Walcott to place this animal amongst the Annelida. They are re-interpreted as gill bars throughout an extended pharynx. This specimen also shows the posterior bilobed structure (USNM 57630). (D, E) The proboscis, collar, trunk and gill bars are all apparent (USNM 57631). (F) The proboscis, circum-collar ridge and gill bars are extremely pronounced (USNM 57632). Direct light images: A, C, D, F; polarized light images: B, E. For acronyms, see Fig. 1. Scale bars: 10 mm...... 76

Appendix 2. Oesia disjuncta from the Burgess Shale (all specimens from Marble Canyon except

USNM 203001, 203033 (C–E – Walcott Quarry). (A–B) Specimen showing the proboscis, gill bars, and exceptional preservation of the posterior bi-lobed structure (B = close-up of framed area in A) (ROMIP63714). (C–E) Nearly complete specimen — anterior region (to the left) missing; pharyngeal gill bars extending almost completely to the terminal end of the trunk and bilobed posterior structure (D, E = close-ups of framed areas in C) (USNM 203001, 203033).

(F–G) Complete specimen showing rounded proboscis, collar, gill bars and bi-lobed posterior structure (ROMIP63709). (H, I) Complete specimen showing rounded proboscis, kidney-heart- stomochord complex, collar, circum-collar ridge, gill bars and posterior structure. A patterning texture in the proboscis suggests possible proboscis muscles (ROMIP63707). Direct light images: A, F, H; Polarized light images: C, D, E, G, I. For acronyms see Fig. 1. Scale bars: A, C,

F, G, =5 mm; B, D, E=1mm; H=10 mm; I=2 mm...... 78

xxvi

Appendix 3. Cluster of Oesia disjuncta from the Burgess Shale (Marble Canyon) preserved on the surface of one large slab (ROMIP63735). O. disjuncta is highly gregarious at the Marble

Canyon, occurring in high abundance across all stratigraphic levels (see also Additional file 7).

(A) Overall slab. (B–C) Close-ups of framed areas in A. The specimen on the left in B is preserved in a decayed tube. The specimen in C shows tripartite body plan, gill bars and kidney- heart-stomochord complex. Polarized light images: A–C. Acronyms see Fig. 1. Scale bars:

A = 10 cm; B, C = 1 cm...... 80

Appendix 4. Close up of a cluster of Oesia disjuncta from the Burgess Shale (Marble Canyon) preserved on the surface of one large slab (ROMIP63736). Specimens show clear tripartite body plan as well as variation in proboscis size and shape. Direct light image: A; Polarized light image: B. For acronyms, see Fig. 1. Scale bars = 1 cm...... 82

Appendix 5. Previously un-figured Oesia disjuncta specimens from the Burgess Shale Walcott's

Quarry – Smithsonian Institution collection. (A–C): Partially dissociated specimen showing nuchal skeleton and gill bars (B and C = close ups of framed areas in A) (USNM 509815) – see also Fig. 1d–e. (D–F) Complete specimen with proboscis slightly dissociated from the rest of the body. The thin element at the posterior of the proboscis is likely the nuchal skeleton. Gill bars are also visible, and the posterior structure is preserved laterally (F = close up of framed area in

E) (USNM 202440). (G–I) Complete specimen possibly showing the nuchal skeleton (USNM

202145, 203031). Direct light images: D, G; Polarized light images: A–C, E, F, H, I. For acronyms, see Fig. 1. Scale bars: A, D, E: 5 mm; B, C: 1 mm; F, I: 2.5 mm; G, H:

10 mm...... 84

xxvii

Appendix 6. Stratigraphic variation in abundance of Oesia disjuncta and Margaretia dorus in the

Marble Canyon paleocommunity. Bars and diamonds indicate numerical abundances of each taxon across 10 cm stratigraphic bins. Coloured bars indicate the percentage of the total number of specimens found within that bin from O. disjuncta and M. dorus (total community size estimates from 2012 and 2014 field collections). Numbers next to the bars indicate the number of occurrences of Oesia preserved inside of Margaretia observed within that bin. Stratigraphic levels on the vertical axis represent negative meters from the boundary between the Eldon and

Stephen Formations as a reference point...... 86

Appendix 7. Detailed imagery of ROMIP63738 (see also Fig. 2d, e). Reflective areas in direct light tend to appear black using polarized light, emphasizing for example the kidney-heart- stomochord complex and the posterior structure (B, D are close-ups of framed areas in A and C).

Direct light images: A, B; polarized light images: C, D. For acronyms see Fig. 1 and Fig. 2. Scale bars: A, C: 10 mm; B, D: 5 mm...... 88

Appendix 8. Additional specimens of Oesia disjuncta preserved inside branching tubes from the

Burgess Shale (Marble Canyon). The worms show a typical high degree of reflectivity in contrast with the surrounding tubes. (A, B) ROMIP63712. (C–E) ROMIP63708. Direct light images: A, D; polarized light images: B, C, E. Br1: branch 1, Br2: branch 2, other acronyms see

Fig. 1 and Fig. 2. Scale bars: A, B, E: 5 mm, C, D = 10 mm...... 90

Appendix 9. Morphology of Margaretia dorus from the Burgess Shale (Raymond Quarry: A, C,

D; Marble Canyon: E; Utah: B; Monarch Cirque: F). (A) Close-up of terminal end showing

xxviii evidence of limited breakage and variability in pore sizes and shapes. The margins of the pores are upraised, giving the tube a semi-corrugated appearance when viewed laterally (ROM

911390). (See also Fig. 3a.) (B) Two specimens from the Spence Shale with possible rounded terminal ends (see arrows) (ROMIP59635). (C, D) Specimen showing a central hole at one end, presumably representing the insertion of a more or less perpendicular branching tube (ROM

63739). (E) Specimen showing a circular structure corresponding to a node of bifurcation similar to ROMIP63739 (ROMIP63706), with a possible rounded terminal end (see arrow). (F)

Complete specimen also illustrated in Fig. 3g–h(framed area see Fig. 3h). This branching tube shows that pore density decreases near the node of bifurcation, that pore shape varies from rhomboid to more ellipse shaped, and that individual fibres are long and continuous through large sections of the tube (ROM 63716). Direct light images: A, B, E; polarized light images: C,

D, F. Te: terminal end; for other acronyms, see Fig. 3. Scale bars: A, E: 5 mm, B, C, D, F:

10 mm...... 92

Appendix 10. Major morphological differences between the two Cambrian, tubicolous enteropneusts Spartobranchus tenuis (top box) and Oesia disjuncta (middle box): (1) S. tenuis is thin and elongate and the trunk much more variable in width compared to O. disjuncta, which is stout and does not vary in width across the length of the trunk; (2) the pharyngeal gill bars are restricted to approximately 10–20 % of the total trunk length [10] in S. tenuis, but extend approximately 80 % of the total trunk length in O. disjuncta; (3) S. tenuis possesses an esophageal organ while O. disjuncta does not; (4) S. tenuis has a bulbous terminal structure while O. disjuncta has a claw-shaped terminal apparatus; also the tube of S. tenuis has an externally corrugated but smooth texture with no evidence of pores or openings (bottom box;

xxix

ROMIP94189, while the tube of O. disjuncta is fibrous, much larger and has helicoidally arranged openings of variable sizes (not illustrated in this figure). Scale bars for the line drawings: 1 cm, scale bars for the tube: 1 mm...... 94

Appendix 11. Biogeographical distribution of the fossil Margaretia and index of specimens used in this study. Reference numbers can be found in the main text...... 96

Chapter 3

Appendix 1. Kootenayscolex barbarensis paratype (ROMIP64389) from the Burgess Shale

(Marble Canyon locality). A: full view of specimen (posterior missing). B-C: elemental maps of full specimen, enriched areas are brighter; B: carbon map shows that carbon tends to be associated with internal features (i.e. gut, nervous tissues and coelomic cavities within the palps; see close up) with the exception of the parapodial vascularized tissues. C: elemental maps for 8 elements. Note non-overlapping Ca/P and Mg/Fe preservation suggesting different diagenetic processes in the parapodia + palps and gut, respectively, presumably as a result of original variations in tissue composition in the different organs. D-E: detail of head region showing putative internal structures under different lighting conditions, wet and cross-polarized (D) and dry and direct light (E). F-G: cathodoluminesence images highlighting continuity of phosphatized areas from the base of the palps to the base of parapodia (G). Scale bars = 1 mm.

Related to Figures 1,2 and 3...... 122

Appendix 2. Kootenayscolex barbarensis from the Burgess Shale (Marble Canyon locality). A: full image of paratype (ROMIP64393) preserved in dorsal-ventral orientation and showing

xxx parapodial morphology as well as extensive gut infill; the head of a second unprepared specimen is also visible on the same slab. B: image of holotype (ROM 64388) under wet conditions. C-D: full images of paratypes (ROMIP64394 – C; ROMIP64397 – D) showing bundles of chaetae preserved only at the anterior end. E-F: paratype (ROMIP64395); E: full image of specimen before preparation. F: close up of framed area in S2E after preparation showing wide, fanned orientation of prostomial chaetae and base of median antennae merging at the level of base of palps within the protostomium. G: close up of paratype (ROMIP62972) showing variations in orientations of the chaetae as a result of variations in angle of burial, and phosphatized areas at the base of both noto- and neuropodia. All images using cross-polarized light. Superimposed images in D of both part and counterpart using Apply Image and overlay blending mode in

Adobe Photoshop CS6 (see thin dashed line for limits between the two parts). Scale bars = 1 mm. Related to Figures 1,2 and 3...... 124

Appendix 3. Kootenayscolex barbarensis from the Burgess Shale (Walcott Quarry) and other

Cambrian polychaetes from the Burgess Shale showing putative peristomial parapodia and chaetae. A-D: single slab showing two relatively complete specimens (ROMIP57190.1-2). B: close up of framed specimen in S3A with elongate palps. C: close up of framed specimen in S3A showing median antenna and dark patches at the base of parapodia reminiscent of the Marble

Canyon specimens. D: counterpart of S3C. E-F: most complete specimen showing both palps, median antenna and chaetae (ROMIP64399). F: close up of anterior portion of S3E. G-H:

Canadia spinosa (lectotype USNM57654), overall view; H: close up of head showing that the first parapodia and chaetae occur posterior to the prostomium. I-K: Burgessochaeta setigera; I-J: previously unfigured specimen (ROMIP64398) showing that the first parapodia and chaetae also

xxxi occur posterior to the prostomium, as in C. spinosa (see detail in J). K: lectotype of B. setigera

(USNM57670) which resembles the proposed morphology of Phragmochaeta canicularis, as a result of both ends of the protostomium being buried into the matrix. Note: posterior end almost indistinguishable from anterior end. Polarized light imagery (A-F, H-J); direct light imagery (G,

K). Scale bars: A=5 mm; B-F: 1 mm; G, H= 2 mm; I, K = 1 mm; J = 0.5 mm. Related to Figures

1,2 and 3...... 126

Appendix 4. Strict consensus parsimony analysis using our updated character matrix to reflect:

(1) the description of Kootenayscolex barbarensis, and (2) re-interpretations of head anatomy of

Canadia spinosa, Burgessochaeta setigera, and Phragmochaeta canicularis (see Star Methods).

The four aforementioned Cambrian taxa form a polytomy with all crown group Annelida

(excluding sipunculids), Guanshanchaeta felicia, and Pygocirrus butyricampum. This large polytomy is weakly supported. Numbers above nodes are Jackknife support values. Numbers below nodes are Bremer support values. Related to Figure 4...... 128

Appendix 5. Character matrix used in all phylogenetic Analyses...... 130

Appendix 6. Figured Specimens of Kootenayscolex barbarensis- Nanglu and Caron 2017...... 140

Chapter 4

Appendix 1. Cluster analyses of the Marble Canyon bedding assemblages using different thresholds for minimum specimen number/bedding assemblage. To produce cumulative bins,

xxxii successive adjacent bedding assemblages were summed to reach a minimum specimen number of 300 before being included in the cluster analysis...... 210

Appendix 2. Burgess Shale species composition data matrix...... 212

Appendix 3. Burgess Shale ecological composition data matrix...... 245

xxxiii

GENERAL INTRODUCTION

The Cambrian Explosion, the name given to the rapid appearance of nearly all major body plans in the fossil record during the Cambrian period, is one of the most significant biological events in Earth’s history (Erwin et al. 2011). This first great radiation of animal life lasted between approximately 542 million years ago (mya) to 505 mya, as evidenced by

Cambrian fossil deposits scattered throughout the globe (Gaines 2014). However, the vast majority of these sites provide evidence only of what are commonly referred to as “small, shelly fauna”; that is, only the most recalcitrant components of skeletonizing animals such as brachiopods, mollusks and trilobites. While useful for biostratigraphy, biogeography, and for some aspects of the evolutionary history of their respective groups, these fossil assemblages lack soft tissue preservation, so are limited in their contributions to: (i) our understanding of the vast majority of animals which possess exclusively more labile tissues (Sansom et al. 2010; Nanglu et al. 2015), (ii) the fine anatomical details of even those taxa which possess harder structures

(Briggs and Kear 1993; Klompmaker et al. 2017), and (iii) the total diversity and structure of animal communities at the dawn of macroscopic animal life (Caron and Jackson 2006; Laflamme et al. 2013).

While there are many fossil deposits around the world from many geologic periods which display exceptional preservation of soft tissues (Muscente et al. 2017), few are as celebrated as the middle-Cambrian Burgess Shale, a 508-million-year-old geological unit within the Stephen

Formation in the Canadian Rockies (Walcott 1910). Since its discovery in 1909 by Charles

Doolittle Walcott, the Burgess Shale has continued to provide invaluable insights into the early evolution of nearly all major metazoan phyla. These studies began with the fossil material

xxxiv collected from the Phyllopod Bed member of the eponymous Walcott Quarry, an informal 2- metre stratigraphic sub-unit of the Walcott Quarry, and even a cursory review of the subsequent papers published on the Phyllopod Bed fauna will reveal the key role these taxa play in illustrating the origins of groups as disparate as sponges(Botting and Muir 2017), comb jellies

(Conway Morris and Collins 1996), arthropods (Briggs 1977; Briggs and Whittington 1985;

Whittington and Briggs 1985), mollusks (Smith and Caron 2010; Smith 2012), annelids (Conway

Morris 1979a), and vertebrates (Morris 2008; Morris and Caron 2012).

The fossil assemblages of the Burgess Shale have not, however, been studied to the same extent as holistic animal communities. While this is partly due to the obvious and disproportionate importance of Burgess Shale fossils for understanding animal evolution (and more broadly, the evolution of body plans as a whole), the rarity of the discovery of new localities that equal the original Walcott Quarry in both fossil richness and species diversity has historically constrained paleocommunity analyses. That being said, considerable strides have been made using the fossils of the Burgess Shale to infer the ecological dynamics of the animals that inhabited the sea floor of the middle Cambrian.

WALCOTT QUARRY AND THE FIRST COMMUNITY ANALYSES

The first true quantitative analysis of a soft-bodied Cambrian paleocommunity was the seminal monograph by Conway Morris (1986), which built upon the broad description of the

Phyllopod Bed community he outlined in 1979 (Conway Morris 1979b). In this study, the stark differences between the resolution of data that could be provided by typical small shelly biota and a soft-tissue-preserving Lagerstätte were laid bare; only 14% of all genera present among the

xxxv

Phyllopod Bed taxa were likely to be preserved in an exclusively small shelly assemblage, and perhaps as little as 2% of the total number of specimens.

While arthropods were unsurprisingly the most abundant taxon in the Phyllopod Bed community, the robust quantitative study quickly differentiated the Walcott Quarry from other

Cambrian localities. The most abundant phyla at the quarry also included the almost entirely soft-bodied hemichordates and priapulids, as well as sponges which, without soft-tissue preservation, would be present only in the form of disarticulated spicules (Conway Morris 1986).

Polychaetes also formed a significant component of the generic diversity, and rarer soft-bodied animals such as Odontogriphus, Pikaia, and the various taxa grouped under the term Miscellanea

(a term introduced by Conway Morris (1985) to describe taxa that, at the time, he suspected may have constituted entirely new phyla) rounded out a much more diverse community than the more common trilobite- and brachiopod-heavy picture of the Cambrian which was dominant at the time. This more comprehensive view also lent itself to a richer study of the ecological modes and niche structure of the Phyllopod Bed. For this purpose, the diversity revealed by soft-tissue preservation proved invaluable. The vast majority of arthropods, and particularly those known solely by their exuviae or isolated elements, were assumed to be epibenthic vagrant animals.

Brachiopods and sponges could justifiably be assumed to be epibenthic and sessile; these two factors together led to a view of Cambrian animal life almost entirely dominated by the epibenthos. The soft-bodied animals provided a window into life both within the benthos—the worms in particular evidenced an active infaunal community—and above it into the nektonic or

‘pelagic’ realms.

Through this study, Conway Morris’s work changed the perception of Cambrian ecosystems from being benthos-dominated with a relatively simple trophic structure when

xxxvi compared with a contemporary marine setting, to exhibiting a complex interplay among animals inhabiting a multitude of familiar marine niches (Bambach et al. 2007). Simultaneously, Conway

Morris’s study emphasized the irreplaceable value of soft-tissue preservation not only for understanding the Cambrian context of early animal evolution, but also of the “evolution” of ecosystems and the relevance of the Burgess Shale for interpreting how ecological communities have changed over geologic time.

Despite its importance, however, this study was not without limitations. The first was that the descriptions of diversity and niche occupation outlined above relied on the Phyllopod Bed being treated as a time-averaged assemblage. That is to say, all specimens were included in a single bulk dataset to be analyzed together irrespective of their stratigraphic positions relative to each other within the Phyllopod Bed sub-unit of the Walcott Quarry. This bulk-scale time averaging was a necessary result of the manner in which the original stratigraphic data was collected by the Geological Society of Canada; multiple turbidites were included within the same vertical sub-units, precluding an analysis of community composition at the individual bedding horizon level of resolution. While unavoidable, this type of bulk analysis relegated our view of the Phyllopod Bed fauna to a monolithic and temporally static community of animals, albeit with some allowance for minor fluctuations in relative abundances throughout the stratigraphic unit.

As Conway Morris put it (1986, page 2) (Conway Morris 1986), “while the Phyllopod Bed fauna was probably one of several benthic communities (possibly intergrading) from the area, its gross faunal composition is representative of the overall faunal diversity of this basin.”

The second limitation was the unique position held by the Walcott Quarry as an unparalleled exemplar for the description of a Cambrian paleocommunity. With no other soft- bodied locality having been analyzed at the time with same degree of quantitative rigour, there

xxxvii was a dearth of acceptable comparators for the Phyllopod Bed community, naturally leading to its ecological structure being expanded to a cosmopolitan model of life during the middle

Cambrian. This view was partially supported by paleoenvironmental data reconstructing the basin in which the Phyllopod Bed community lived as facing the open sea with minimal physical boundaries, suggesting that “the Phyllopod Bed was open to migration from distant parts of the

Earth” (Conway Morris 1986, page 3). Further, the fact that nearly all Cambrian North American genera of trilobites and brachiopods were represented in the Walcott Quarry, combined with the wide distribution of these taxa, led Conway Morris to conclude that, “the position of the Burgess

Shale, with its indigenous deep-water faunas, argues for a cosmopolitan aspect” (1986, page 3).

CHANGING VIEWS OF CAMBRIAN PALEOECOLOGY

Building upon Conway Morris’s work, Caron and Jackson re-examined the Walcott

Quarry material to assess both the taphonomic setting of the Greater Phyllopod Bed (GPB)—an expanded geological sub-section 7 metres in total thickness, as opposed to the 2 metre sub- section used in Conway Morris’s study—as well as the diversity of the locality within a modern phylogenetic context (Caron and Jackson 2006, 2008). This expanded stratigraphic sampling effort, combined with the incorporation of fine-scale stratigraphic measurements of species occurrences into the community dataset and more nuanced statistical techniques than had been applied previously, have proven invaluable in broadening our understanding of community ecology during the middle Cambrian.

The most crucial finding of these studies was that when stratigraphic data was incorporated, the Phyllopod Bed community was far from static in either its relative species

xxxviii abundances or its representative ecological modes. In terms of simple metrics of species richness and evenness, there is no observable unidirectional stratigraphic trend. Instead, stratigraphic sub- units of high and low diversity are interspersed throughout the GPB, indicating pulses of changing diversity over time (Caron and Jackson 2008). Similarly, the actual species composition of the sub-units does not constitute a smooth gradient of changing taxa through time, but, rather, there are four major observable “groups,” each with its own assemblage of associated species that occur and re-occur in disjunct temporal order.

These studies definitively demonstrated that the assembly of communities at the best studied Cambrian fossil locality was temporally dynamic. Consequently, we can expect a considerable loss of valuable ecological information when analyzing Cambrian fossil deposits as bulk assemblages. Doing so runs the risk of making false, or at least un-nuanced, comparisons between a locality studied as a bulk assemblage and other community datasets, both paleontological and extant. They also largely rectified the first limitation of Conway Morris’s work (bulk-assemblage-level time averaging) by adding a finer scale temporal resolution to the

GPB. The second limitation to Conway Morris’s studies, the need for comparable Cambrian datasets from further afield, would be partially remedied by paleocommunity analyses of two other Burgess Shale localities: the Raymond Quarry and the Tulip Beds.

In an unpublished MSc thesis, Devereux (2001) examined the fossils from the Burgess

Shale Raymond Quarry locality. The specimens collected from this locality have a similar degree of stratigraphically-detailed measurements as those from the Phyllopod Bed, permitting fossils to be binned into 10 cm sub-units for analysis of community changes over time. Furthermore, due to its position 22 metres directly above Walcott Quarry, it is the best constrained site with reference to the stratigraphic position of the well-studied Phyllopod Bed fauna. The patterns of

xxxix species abundance over time largely recapitulate the major findings from the GPB, with significant fluctuations in diversity and faunal composition occurring in pulses rather than as a smooth gradient reflective of stratigraphic order. Also, while most taxa are shared between the

GPB and Raymond Quarry assemblages, their relative abundances differ significantly; the sponge Choia, leanchoilids, the priapulid Ottoia and the still enigmatic Pollingeria together making up roughly 68% of the total abundance of the Raymond Quarry.

While these data is valuable, the Raymond Quarry study did not include significant comparisons with the GPB fauna as a bulk assemblage, nor were the patterns of faunal turnover of GPB fauna known at the time, since it pre-dated the re-investigations of the GPB fossils by seven years (Caron and Jackson 2006, 2008). Furthermore, the Raymond Quarry fauna does not necessarily represent an entirely different locality from the Walcott Quarry with regard to the pre-mud slide environment being sampled. In this way, it may be more accurate to see the

Raymond Quarry as an extension of the Walcott Quarry that is disjunct temporally rather than geographically. It therefore represents another valuable source of species abundance data for the

Cambrian that is informative at an ecologically relevant timescale, but it does not satisfy the need for an independent sampling effort at a novel locality.

The paleocommunity analysis of the Tulip Beds, located on Mt. Stephen, does satisfy this requirement, representing an independent locality (O’Brien and Caron 2015). One of the best- sampled Burgess Shale sites, it includes over 9,000 fossils, notably the cryptic “tulip animal”

Siphusauctum gregarium, from which the locality draws its name (O’Brien and Caron 2012).

Furthermore, due to its stratigraphic position as part of the Campsite Cliff Shale Member of the

Stephen Formation, the age of Tulip Beds is well established as older than the Walcott Quarry and Raymond Quarry, which are both part of the younger Walcott Quarry Member of the

Stephen Formation (Fletcher and Collins 1998; O’Brien et al. 2014). Quantitative analyses demonstrate that many broad patterns are shared between the Tulip Beds and GPB, despite the geographic distance between them. Unsurprisingly, arthropods are the most abundant and diverse group at both sites, with sponges being the second-most abundant. Ecologically, the communities of both fossil localities are epibenthos-dominated. However, there are also major differences between these communities. Most obviously, the faunal compositions differ significantly, with the Tulip Beds fauna sharing only about half of its genera with the GPB. Also, in a manner similar to the Raymond Quarry, even shared species vary drastically in their relative abundances between sites. Perhaps the most striking example of this phenomenon can be found in the presence of the iconic Anomalocaris canadensis. Often considered to be the top predator of the

Cambrian seas, the mention of Anomalocaris immediately conjures images of its silhouette gliding over hapless trilobites, claws at the ready. Yet, at the Walcott Quarry, only a single specimen of Anomalocaris has been collected. In stark contrast, Anomalocaris is the fourth most abundant taxon at the Tulip Beds, making up 4% of the community by abundance (O’Brien and

Caron 2015). Combined with the locality-defining Siphusauctum and a number of as yet undescribed taxa, the Tulip Beds demonstrate that while some very general aspects of community structure may have been recurrent among Cambrian localities, detailed observations reveal a significant degree of local specificity. Crucially, however, the Tulip Beds fauna represents another bulk assemblage, as the fossils are almost all talus material, and thus cannot be integrated into stratigraphic occurrence data (O’Brien and Caron 2015).

To summarize, while the studies of both the Raymond Quarry and Tulip Beds have contributed significantly to our knowledge of community ecology in the middle Cambrian, neither of these localities satisfies both of the crucial criteria for an ideal point of comparison

xli with the GPB. While the Raymond Quarry yields detailed stratigraphic data for its constituent specimens, its position directly above the Walcott Quarry hampers its ability to be used as an independent and geographically distinct comparator for both diversity and patterns of community turnover. The Tulip Beds, while geographically further afield and representing a significantly different faunal assemblage, does not possess the stratigraphic resolution required for meaningful comments to be made regarding patterns of diversity change at the same temporal scale as the

GPB community. Indeed, for a fossil locality to meet both of these criteria, while also presenting the same exceptional quality of soft-tissue preservation as the Walcott Quarry, is such a stringent threshold that it would not be met for over 100 years since Charles Doolittle Walcott’s discovery of the Burgess Shale itself.

THE MARBLE CANYON FOSSIL LOCALITY

The Marble Canyon fossil site was discovered in 2012 by a Royal Ontario Museum-led research team (Caron et al. 2014). Located in Kootenay National Park, 40 km southwest of the

Walcott Quarry, the importance of this new site was easily recognizable. Both the exceptional quality of fossil preservation as well as the raw density of fossil occurrences drew immediate comparisons with the Walcott Quarry, but, astonishingly these were not the most surprising finds. Preliminary analyses of approximately 3,000 specimens suggested that 22% of the species found at Marble Canyon were entirely new, underscoring the absolute necessity of discovering new localities to build a comprehensive picture of middle Cambrian species diversity. Also interestingly, the contact point between the Stephen Formation and the overlying Eldon

xlii

Formation suggested that the Marble Canyon may be the youngest discovered Burgess Shale site to date (Caron et al. 2014).

The wealth of new data that these fossils represented led to two subsequent sampling efforts in 2014 and 2016, bringing the total collected fossil material to over 20,000 specimens. A core crew returning each trip ensured that stratigraphic data was taken for each of these specimens and was correlated across collection years to ensure consistency of measurement.

Living up to the initial excitement of its discovery, Marble Canyon now stands as one of the most robustly sampled Cambrian paleocommunities, with its constituent taxa contributing to a series of new fossil descriptions and important redescriptions over the last four years (Morris and

Caron 2014; Aria and Caron 2015, 2017; Aria et al. 2015; Nanglu et al. 2016; Moysiuk et al.

2017; Nanglu and Caron 2018). The analyses I have undertaken as part of my doctoral research demonstrate that the Marble Canyon is an ideal site to compare with the Walcott Quarry in terms of fine resolution stratigraphic detail and as an independent, geographically distinct fossil locality with its own unique biota. Further, the possibility of the Marble Canyon representing one of the youngest studied Burgess Shale fauna localities can be investigated through the integration of more stratigraphically-constrained localities such as the Raymond Quarry and Tulip Beds to provide an unprecedented view of faunal turnover throughout the Burgess Shale. Taken in concert with previous paleocommunity analyses, Marble Canyon stands to contribute significantly to our understanding of ecological structure and faunal assembly among some of the earliest, complex animal communities.

xliii

OBJECTIVES OF THIS THESIS

The goal of my dissertation is to describe the structure and diversity of the Marble

Canyon paleocommunity and to use these insights to reassess the overall structure of the Burgess

Shale as a temporal “metacommunity” by integrating this new dataset with those of the GPB and

Raymond Quarry. This goal has been met through 3 complementary avenues:

i) a decay study using an extant analogue of the hemichordate Oesia disjuncta, a

taxon whose great abundance is characteristic of the Marble Canyon

paleocommunity, to assess potential taphonomic bias in my community structure

interpretations (Chapter 1);

ii) descriptions and redescriptions of key enigmatic taxa that constitute significant

components of the Marble Canyon fauna, without which it would be impossible to

analyze both patterns of temporal turnover in species composition and trophic

structure of the community (Chapters 2 and 3); and

iii) a quantitative analysis of the Marble Canyon fauna through a series of 10 cm sub-

units collected within a ca. 4 metre-thich interval at one locality (Chapter 4).

Comparisons between Marble Canyon and other Burgess Shale sites (Walcott

Quarry, the Tulip Beds and the Raymond Quarry) allow a broader understanding

of spatial and temporatl variations of the Burgess Shale paleocommunity.

xliv

CHAPTER SUMMARY

CHAPTER 1

This study presents a taphonomic analysis of the rate and sequence of decay of morphological characters post-mortem in three species of acorn worm (Enteropneusta,

Hemichordata) (Nanglu et al. 2015). These data are used to assess the phylogenetic affinities of the recently-described hemichordate Spartobranchus tenuis (Caron et al. 2013). Next, these data are used to assess possible taphonomic biases in my community dataset for the Marble Canyon in Chapter 4, specifically by using the quality of preservation of the abundant hemichordate

Oesia disjuncta as an index for the loss of species-presence information due to decay. Oesia disjuncta is uniquely well suited for this task, as it is both abundant (the third most abundant taxon) and stratigraphically widespread (present in every 10 cm temporal sub-unit analyzed) in the quarry.

The results of this chapter also serve as a repudiation of the stemwards-slippage hypothesis as a ubiquitous factor influencing the interpretation of fossil morphology (Sansom et al. 2010, 2011). In hemichordates, the decay of morphological characters does not seem to share any correlation with the importance of the character in question in informing phylogenetic analyses. The sequence of character loss is more accurately, and more logically, a consequence of a combination of: i) the positioning of the structure relative to the external environment; ii) the size of the structure; and iii) the composition of the structure. This debate surrounding the extent to which decay biases our interpretations of the anatomy and phylogenetic affinities of fossil taxa remains one of the most contentious in palaeontology and is particularly relevant to early

xlv

Palaeozoic fossils, many of which are part of the stem lineages of major modern clades

(Donoghue and Purnell 2009; Parry et al. 2018; Purnell et al. 2018). This chapter was originally published in the journal Paleobiology.

CHAPTER 2

This study presents a redescription of the enigmatic fossil Oesia disjuncta (Nanglu et al.

2016). While both rare and poorly preserved at the Walcott Quarry, Oesia disjuncta is among the most abundant taxa at Marble Canyon. Multiple phylogenetic affinities have been proposed for this taxon over the last 100 years, with no consensus (Walcott 1911; Lohmann 1920; Szaniawski

2005, 2009; Conway Morris 2009). A full description utilizing the new collections from Marble

Canyon has revealed that Oesia is,in fact, a hemichordate with an anatomy utterly unlike any modern analogue.

The presumed alga Margaretia dorus is similarly abundant at Marble Canyon. My analysis has also revealed that Margaretia dorus is actually a tube-shaped dwelling constructed by Oesia disjuncta, which has significant implications for the origins of tube-building in the phylum Hemichordata. Just as importantly, this shifts Margaretia from its previously-accepted position as a macroscopic alga (Satterthwait 1976; Conway Morris and Robison 1988) to the tube of another animal. We have identified Aldanophyton from the Siberian Sinsk biota and a currently undescribed fossil from Yunnan, China as species of Margaretia (Ivantsov et al. 2005;

Hu et al. 2010). Combined with the wide distribution of Margaretia at other Laurentian localities in North America, this new understanding of its nature significantly expands the biogeographic range and ecological importance of the hemichordates in the early Palaeozoic.

xlvi

The studies of these two taxa were crucial for analyzing the diversity and ecological structure of the Marble Canyon: Oesia disjuncta is now the preeminent suspension feeding organism in many temporal sub-units, and the fact that Margaretia dorus does not represent an abundant macroscopic alga suggests that the Marble Canyon fauna was not restricted to the photic zone. This chapter was originally published in the journal BMC Biology.

CHAPTER 3

This study presents a description of a new species of polychaete from the Marble Canyon quarry, Kootenayscolex barbarensis. Annelids are notably rare in the early Palaeozoic fossil record, with only eight taxa previously described from the Cambrian. A fossil from Marble

Canyon initially identified as a species of Burgessochaeta actually represents the first new polychaete discovered from the Burgess Shale in nearly 40 years. It is also among the most abundant taxa at Marble Canyon and, in fact, is among the most abundant and well-preserved annelids in the entire fossil record. This study was necessary for the completion of the palaecommunity analysis, as this new and previously undescribed species represents one of the most ecologically significant taxa within the richest fossiliferous assemblages in the Marble

Canyon (described below).

The results of this chapter have also led to the formulation of a new hypothesis for the origins of the modern annelid head, incorporating parallels with the developmental biology of a number of extant polychaetes. More specifically, we describe how similar developmental pathways as those controlling the fusion of the first chaetigerous segment of the extant nereid polychaetes (Mazurkiewicz 1975, 2009; Fischer et al. 2010) with the peristomium—a process

xlvii termed cephalic metamorphosis—and the re-absorportion of transient parapodia and chaetae into the body of magelonid polychaetes (Wilson 1982) may have produced the unique head morphologies found among Kootenayscolex barbarensis and other Cambrian annelids (i.e.

Canadia spinosa and Burgessochaeta setigera). This hypothesis contradicts the ‘metameric head’ hypothesis that has recently been proposed for the early evolution of the Annelida, which posits a homonomously segmented and entirely bristled morphology (Parry et al. 2015). This chapter was originally published in the journal Current Biology.

CHAPTER 4

This study presents an analysis of the paleocommunity diversity and structure of the

Marble Canyon fauna. It includes analyses of the taxonomic and ecological diversity of the community both as a bulk assemblage and also partitioned into 10 cm sub-units to assess patterns of community change over time. Quantitative analyses include the use of rarefaction curves, species abundance curves, ordination methods (particularly correspondence analysis), cluster analysis, beta diversity indices (ie. Morisita-Horn index). The Marble Canyon dataset is then integrated with the already-published Greater Phyllopod Bed and Tulip Beds datasets as well as newly collected data from the 2.2 m interval of Raymond Quarry strata and analyzed using ordination and clustering methods.

Faunal turnover patterns reveal that the Marble Canyon fauna was temporally dynamic in both its ecological and taxonomic composition. Compared with the Walcott Quarry, which possesses four major taxonomic assemblage types that recur in temporally disjunct order (Caron and Jackson 2008), the Marble Canyon fauna changed in composition along a smoother gradient

xlviii that possessed only two major points of inflection. Multivariate analyses of all three localities

(Walcott Quarry, Raymond Quarry, and Marble Canyon) show that, while bedding assemblages from each locality are taxonomically dissimilar from those of other localities, they conform to ecological structures that recur throughout all three sites.

Broadly, the Burgess Shale paleocommunity can be seen to be extremely heterogeneous in its taxonomic composition across the three best-sampled soft-bodied localities. However, each of these localities shares the commonality of a pattern of significant faunal turnover. These two factors together suggest that animal communities during the middle Cambrian were compositionally dynamic at the scale of landscapes, individual localities, and over ecologically relevant timescales.

xlix

REFERENCES

Aria, C., and J. B. Caron. 2015. Cephalic and limb anatomy of a new isoxyid from the burgess

shale and the role of “Stem bivalved arthropods” in the disparity of the frontalmost

appendage. PLoS ONE 10:1–37.

Aria, C., and J.-B. Caron. 2017. Burgess Shale fossils illustrate the origin of the mandibulate

body plan. Nature 545:89–92.

Aria, C., J.-B. Caron, and R. Gaines. 2015. A large new leanchoiliid from the Burgess Shale and

the influence of inapplicable states on stem arthropod phylogeny. Palaeontology 58:629–

660.

Bambach, R. K., A. M. Bush, and D. H. Erwin. 2007. Autecology and the filling of ecospace:

key metazoan radiations. Palaeontology 50:1–22.

Botting, J. P., and L. A. Muir. 2017. Early sponge evolution: A review and phylogenetic

framework. Palaeoworld 27:1–29.

Briggs, D. E. G. 1977. Bivalved arthropods from the Cambrian Burgess Shale of British

Columbia. Palaeontology 20:67–72.

Briggs, D. E. G., and H. B. Whittington. 1985. Modes of life of arthropods from the Burgess

Shale, British Columbia. Transactions of the Royal Society of Edinburgh: Earth Sciences

76:149–160.

Briggs, D. E. G., and A. J. Kear. 1993. Decay and preservation of polychaetes: taphonomic

thresholds in soft-bodied organisms. Paleobiology 19:107–135.

Caron, J.-B., and D. A. Jackson. 2006: Taphonomy of the Greater Phyllopod Bed community,

Burgess Shale. Palaios 21:451–465.

Caron, J.-B., and D. A. Jackson. 2008. Paleoecology of the Greater Phyllopod Bed community,

Burgess Shale. Palaeogeography, Palaeoclimatology, Palaeoecology 258:222–256.

Caron, J.-B., S. C. Morris, and C. B. Cameron. 2013. Tubicolous enteropneusts from the

Cambrian period. Nature 495:503–506.

Caron, J.-B., R. R. Gaines, C. Aria, M. G. Mángano, and M. Streng. 2014. A new phyllopod bed-

like assemblage from the Burgess Shale of the Canadian Rockies. Nature communications

5:3210.

Conway Morris, S. 1979a. Middle Cambrian polychaetes from the Burgess Shale of British

Columbia. Philosophical Transactions of the Royal Society B: Biological Sciences

285:227–274.

———. 1979b. The Burgess Shale (Middle Cambrian) Fauna. Annual Review of Ecology and

Systematics 10:327–349.

———. 1985: Cambrian Lagerstatten. Their distribution and significance. Philosophical

Transactions of the Royal Society B: Biological Sciences 311:49–65.

———. 1986. The community structure of the Middle Cambrian phyllopod bed (Burgess Shale).

Palaeontology 29:423–467.

———. 2009. The Burgess Shale animal Oesia is not a Chaetognath: A Reply to Szaniawski

(2005). Acta Palaeontologica Polonica 54:175–179.

Conway Morris, S., and R. A. Robison. 1988. More soft-bodied animals and algae from the

Middle Cambrian of Utah and British Columbia. University of Kansas Paleontological

Contributions.

Conway Morris, S., and D. Collins. 1996. Middle Cambrian ctenophores from the Stephen

Formation, British Columbia, Canada. Philosophical Transactions of the Royal Society B:

Biological Sciences 351:279–308.

Devereux, M. G. 2001. Palaeoecology of the Middle Cambrian Raymond Quarry fauna, Burgess

Shale, British Columbia. University of Western Ontario Master’s Thesis.

Donoghue, P. C. J., and M. a. Purnell. 2009. Distinguishing heat from light in debate over

controversial fossils. BioEssays 31:178–189.

Erwin, D. H., M. Laflamme, S. M. Tweedt, E. A. Sperling, D. Pisani, and K. J. Peterson. 2011.

The Cambrian conundrum: early divergence and later ecological success in the early history

of animals. Science 334:1091–1097.

Fischer, A. H. L., T. Henrich, and D. Arendt. 2010. The normal development of Platynereis

dumerilii. Frontiers in Zoology 7:1–39.

Fletcher, T., and D. Collins. 1998. The Middle Cambrian Burgess Shale and its relationship to

the Stephen Formation in the southern Canadian Rocky Mountains. Canadian Journal of

Earth Sciences 35:413–436.

Gaines, R. R. 2014. Burgess shale-type preservation and its distribution in space and time.

Palaeontological Society Papers 20:123–146.

Hu, S. X., M. Y. Zhu, M. Steiner, H. L. Luo, F. C. Zhao, and Q. Liu. 2010. Biodiversity and

taphonomy of the Early Cambrian Guanshan biota, eastern Yunnan. Science China Earth

Sciences 53:1765–1773.

Ivantsov, A. Y., A. Y. Zhuravlev, A. V. Leguta, V. a. Krassilov, L. M. Melnikova, and G. T.

Ushatinskaya. 2005. Palaeoecology of the early cambrian Sinsk biota from the Siberian

platform. Palaeogeography, Palaeoclimatology, Palaeoecology 220:69–88.

Klompmaker, A. A., R. W. Portell, and M. G. Frick. 2017. Comparative experimental

taphonomy of eight marine arthropods indicates distinct differences in preservation

potential. Palaeontology 60:773–794.

lii

Laflamme, M., S. A. F. Darroch, S. M. Tweedt, K. J. Peterson, and D. H. Erwin. 2013. The end

of the Ediacara biota: Extinction, biotic replacement, or Cheshire Cat? Gondwana Research

23:558–573.

Lohmann, H. 1920.Oesia disjuncta Walcott, eine Appendicularie aus dem Kambrium.

Mitteilungen aus dem Zoologischen Staatsinstitut und Zoologischen Museum in Hamburg

38:69–75.

Mazurkiewicz, M. 1975. Larval development and habits of Laeonereis culveri (Webster)

(Polychaeta: Nereidae). Biological Bulletin:186–204.

———. 2009. The ontogeny of parapodia and setae in Laeonereis culveri (Webster) (Polychaeta:

Nereididae). Zoosymposia 2:369–377.

Morris, S. C. 2008. A redescription of a rare chordate, Metaspriggina walcotti Simonetta and

Insom, from the Burgess Shale (Middle Cambrian), British Columbia, Canada. Journal of

Paleontology 82:424–430.

Morris, S. C., and J. B. Caron. 2012. Pikaia gracilens Walcott, a stem-group chordate from the

Middle Cambrian of British Columbia. Biological Reviews 87:480–512.

Morris, S. C., and J.-B. Caron. 2014. A primitive fish from the Cambrian of North America.

Nature 512:419–422.

Moysiuk, J., M. Smith, and J.-B. Caron. 2017. Hyoliths are Palaeozoic lophophorates. Nature

541:394–397.

Muscente, A. D., J. D. Schiffbauer, J. Broce, M. Laflamme, K. O’Donnell, T. H. Boag, M.

Meyer, A. D. Hawkins, J. W. Huntley, M. McNamara, L. A. MacKenzie, G. D. Stanley, N.

W. Hinman, M. H. Hofmann, and S. Xiao. 2017. Exceptionally preserved fossil

assemblages through geologic time and space. Gondwana Research 48:164–188.

liii

Nanglu, K., and J.-B. Caron. 2018. A new Burgess Shale polychaete and the origin of the annelid

head revisited. Current Biology 28:319-326.

Nanglu, K., J.-B. Caron, and C. B. Cameron. 2015. Using experimental decay of modern forms

to reconstruct the early evolution and morphology of fossil enteropneusts. Paleobiology 41.

Nanglu, K., J.-B. Caron, S. Conway Morris, and C. B. Cameron. 2016. Cambrian suspension-

feeding tubicolous hemichordates. BMC Biology 14:9.

O’Brien, L. J., and J. B. Caron. 2012. A new stalked Filter-Feeder from the Middle Cambrian

Burgess Shale, British Columbia, Canada. PLoS ONE 7.

O’Brien, L. J., and J. Caron. 2015. Paleocommunity analysis of the Burgess Shale Tulip Beds ,

Mount Stephen , British Columbia: comparison with the Walcott Quarry and implications

for community variation in the Burgess Shale. Paleobiology:1–27.

O’Brien, L. J., J.-B. Caron, and R. R. Gaines. 2014. Taphonomy and depositional setting of the

Burgess Shale Tulip Beds, Mount Stephen, British Columbia. PALAOIS 29:309–324.

Parry, L., J. Vinther, and G. D. Edgecombe. 2015. Cambrian stem-group annelids and a

metameric origin of the annelid head. Biology Letters 11:1–5.

Parry, L. A., F. Smithwick, K. K. Nordén, E. T. Saitta, J. Lozano-Fernandez, A. R. Tanner, J. B.

Caron, G. D. Edgecombe, D. E. G. Briggs, and J. Vinther. 2018. Soft-bodied fossils are not

simply rotten carcasses – toward a holistic understanding of exceptional fossil preservation.

BioEssays 40:1–11.

Purnell, M. A., P. J. C. Donoghue, S. E. Gabbott, M. E. McNamara, D. J. E. Murdock, and R. S.

Sansom. 2018. Experimental analysis of soft-tissue fossilization: opening the black box.

Palaeontology 61:317–323.

Sansom, R. S., S. E. Gabbott, and M. A. Purnell. 2010. Non-random decay of chordate

liv

characters causes bias in fossil interpretation. Nature 463:797–800.

Sansom, R. S., S. E. Gabbott, and M. A. Purnell. 2011. Decay of vertebrate characters in hagfish

and lamprey (Cyclostomata) and the implications for the vertebrate fossil record.

Proceedings of the Royal Society B: Biological Sciences 278:1150–1157.

Satterthwait, D. F. 1976. Paleobiology and Paleoecology of Middle Cambrian Algae from

Western North America.

Smith, M. R. 2012. Mouthparts of the Burgess Shale fossils Odontogriphus and Wiwaxia:

implications for the ancestral molluscan radula. Proceedings of the Royal Society B:

Biological Sciences 279:4287–4295.

Smith, M. R., and J. B. Caron. 2010. Primitive soft-bodied cephalopods from the Cambrian.

Nature 465:469–472.

Szaniawski, H. 2005. Cambrian chaetognaths recognized in Burgess Shale fossils. Acta

Palaeontologica Polonica 50:1–8.

———. 2009. Fossil Chaetognaths from the Burgess Shale: A Reply to Conway Morris (2009).

Acta Palaeontologica Polonica 54:361–364.

Walcott, C. D. 1910. Cambrian geology and paleontology. Smithsonian Miscellaneous

Collections 53:51–137.

Walcott, C. D. 1911. Cambrian geology and paleontology. Middle Cambrian annelids.

Smithsonian Miscellaneous Collections 57:107–144.

Whittington, H. B., and D. E. G. Briggs. 1985. The Largest Cambrian Animal, Anomalocaris,

Burgess Shale, British Columbia. Phil. Trans. R. Soc. B 309:569–609.

Wilson, D. P. 1982. The larval development of three species of Magelona (Polychaeta) from

localities near Plymouth. Journal of the Marine Biological Association of the United

Kingdom:385–401.

lvi

CHAPTER 1 Using experimental decay of modern forms to reconstruct the early evolution and morphology of fossil enteropneusts

This chapter is formatted for and was originally published in the journal Paleobiology, and is reproduced here with the permission of that journal. Supplemental data are included as appendices.

ABSTRACT

Decay experiments are becoming a more widespread tool in evaluating the fidelity of the fossil record. Character interpretations of fossil specimens stand to benefit from an understanding of how decay can result in changes in morphology and, potentially, total character loss. We performed a decay experiment for the Class Enteropneusta to test the validity of anatomical interpretations of the Burgess Shale enteropneust Spartobranchus tenuis and to determine how the preservation of morphological features compares with the sequence of character decay in extant analogues. We used three species of enteropneust (Saccoglossus pusillus, Harrimania planktophilus and Balanoglossus occidentalis) representing the two major families of

Enteropneusta. Comparisons between decay sequences suggest morphological characters decay in a consistent and predictable manner within Enteropneusta, and do not support the hypothesis of stem-wards slippage. The gill bars and nuchal skeleton were the most decay resistant, whereas the gill pores and pre-oral ciliary organ were unequivocally the most decay prone. Decay patterns support the identification of the nuchal skeleton, gill bars, esophageal organ, trunk and proboscis in Spartobranchus tenuis and corroborate a harrimaniid affinity. Bias due to the taphonomic loss of taxonomically informative characters is unlikely. The morphologically simple harrimaniid body plan can be seen, therefore, to be plesiomorphic within the

1 enteropneusts. Discrepancies between the sequence of decay in a laboratory setting and fossil preservation also exist. These discrepancies are highlighted not to discredit the use of modern decay studies, but underline their non-actualistic nature. Paleo-environmental variables besides decay, such as the timeframe between death and early diagenesis as well as post-mortem transport, are discussed relative to decay data. These experiments reinforce the strength of a comprehensive understanding of decay sequences as a benchmark against which to describe fossil taxa and understand the conditions leading to fossilization.

INTRODUCTION

Decay bias presents a potentially significant impediment to the interpretation of fossils, especially those of soft-bodied organisms preserved in various fossil Lagerstätten. With the onset of decay, taxonomically informative features can be lost or altered. Estimating the amount of decay biases is therefore of particular significance when considering, for example, problematic Cambrian soft-bodied fossils for which multiple disparate interpretations have been offered (Yunnanozoon: Chen et al. 1995; Shu et al. 1996; Chen and Huang 2008; for a review of several case studies, see Donoghue and Purnell 2009). Delineating taphonomic loss from true phylogenetic absence remains a central question in paleontology and this question has received renewed interest in recent years, especially since the introduction of the “stem-wards slippage” concept (Sansom et al. 2010a, 2010b). This concept is based on the theory that taxa are placed erroneously basal due to the underlying bias of many synapomorphic characters to early decay over plesiomorphic ones. However, it remains to be seen if stem-wards slippage is ubiquitous among animal groups beyond vertebrates (Sansom and Wills 2014).

Decay experiments potentially provide a useful tool to estimate taphonomic biases. Such experiments have been used to investigate a number of factors, including: the role of anoxia in the preservation of soft-bodied organisms (Allison 1986), the relative contribution of transport in fossil articulation and completeness (Allison 1988), the degree to which decay alters soft tissues

(Briggs and Kear 1993a) and how such data can be useful to infer taphonomy at the community level (Briggs and Kear 1993b; Caron and Jackson 2006). Decay experiments are also used to infer the physical and chemical composition of extinct taxa and the potential chemical and geologic pathways of preservation (Briggs and Kear 1994; Briggs et al. 1995), including effects of sediment type on rate of decomposition (Butterfield 2013) and the preservation enhancing potential of microbial death masks on organic material (Darroch et al. 2012). The quantification of the cumulative rate of decay of morphological characters over time (Casenove et al. 2011), and creating cross comparisons of decay sequences among phylogenetically important extant proxies to serve as a visual guide for fossil interpretation (Sansom et al. 2013) have also stemmed from decay experiment studies. However, the usefulness of such studies has also been called into question when they are used without a critical reappraisal of the fossil record, including direct and extensive comparisons of data from decay experiments with fossils (Conway

Morris and Caron 2012).

Here, we present the experimental taphonomy and sequence of decay and decomposition of three enteropneust (acorn worm) species from two phylogenetically distant families (Fig. 1).

This research is the most extensive comparison of decay sequences of a single fossil taxon and the first of its kind for acorn worms. The Hemichordata, including the class Enteropneusta, are sister to the Echinodermata, with which they form the superphylum Ambulacraria.

Ambulacrarians are sister taxa to the Chordates and together comprise the deuterostome branch

3 of animal life (Cameron 2005). While these inter-phylum relationships are strongly supported, the phylogeny within the hemichordates has been a matter of historical debate. Much of this debate centers on the relationship between the two hemichordate classes, which are morphologically disparate: the Enteropneusta are solitary and vermiform, whereas the

Pterobranchia are colonial and tube-dwelling. The pterobranchs, primarily the graptolites, are abundant in the fossil record due to the robustness of their tubaria, and have important implications for dating rock sequences (Mitchell et al. 2013; Maletz 2014).

In contrast, findings of putative enteropneusts in the fossil record have been rare (Arduini et al. 1981; Bechly and Frickhinger 1999; Alessandrello et al. 2004), taxonomically contentious

(Szaniawski 2005; Conway Morris 2009; Szaniawski 2009) and descriptively uninformative

(Shabica and Hay 1997). First described in 1911 from the Burgess Shale, Yoho National Park,

Spartobranchus tenuis was recently redescribed as an enteropneust living in tubes and most resembling the extant family Harrimaniidae (Caron et al. 2013; but see Halanych et al. 2013 and

Cannon et al. 2013 for an alternate hypothesis). Three factors make S. tenuis critical to understanding the origin and early evolution of hemichordates. First, its exceptional quality of preservation reveals more aspects of the internal anatomy than any other fossil enteropneust.

Second, its Cambrian origins push back the earliest known instance of enteropneusts in the fossil record approximately 200 million years (the previous earliest known fossil enteropneusts are

Pennnsylvanian). Third, thousands of specimens allow us to make robust cross comparisons of anatomical features and correctly identify anomalous characters which may be artifacts of decay/preservation. Furthermore, the broad morphological similarity of modern taxa to S. tenuis means that acorn worms are ideal candidates for an experimental decay study; direct comparisons of their anatomies may be made without the need to observe decay in a range of

4 extant species to account for different possible morphological aspects of the ancestral form

(Sansom et al. 2013). These factors, when considered alongside the hypothesis that the ancestral deuterostome body plan may have resembled that of an acorn worm (Cameron et al. 2000), underline the need for a comprehensive understanding of the early evolutionary history of the hemichordates (Swalla et al. 2008). The goals of our experiment were therefore to: (1) describe the sequence and rate of morphological character decay of the Enteropneusta, (2) perform direct and exhaustive comparisons with the anatomy of the fossil enteropneust S. tenuis, (3) consider the phylogenetic position of S. tenuis and implications on the early evolution of the

Enteropneusta and inter-relationships within Hemichordata, (4) analyze the strengths and weaknesses of decay experiments for describing fossil anatomy, and (5) use our understanding of enteropneust decay to make inferences regarding the taphonomic setting of the Burgess Shale.

MATERIALS AND METHODS

Saccoglossus pusillus and Harrimania planktophilus (n = 60 and 34, respectively) were collected by KN and CBC in the Cape Beale Inlet, British Columbia, Canada in May and June,

2013, using a shovel and hand sieve. Specimens were immediately transported to the nearby

Bamfield Marine Sciences Centre and were kept alive with continuous circulating seawater for a maximum of two weeks prior to euthanasia. Balanoglossus occidentalis (n = 12) were collected by CBC in July 2013 from Penrose Point State Park, Washington, USA, and shipped in seawater on ice to the Royal Ontario Museum in Toronto. Due to the limited number of specimens available and the frequency of breakage of B. occidentalis specimens during extraction, experiments were conducted on fragmented as well as whole individuals. All damage occurred posterior to the pharyngeal trunk; pharyngeal fragments were placed with corresponding trunk

5 fragments during the experiment. Only completely undamaged specimens of S. pusillus and H. planktophilus were used.

All specimens were euthanized following previous protocols (Cameron 2002), with a prolonged exposure to a 50:50 solution of magnesium chloride (MgCl2, 75 mg per ml) and artificial seawater (Instant Ocean brand artificial seawater, specific gravity = 1.02). In enteropneusts, determining time of death, and therefore the theoretical beginning of decay, is difficult. In the absence of a brain, or any true centralized nervous system, “death” can be an ambiguous term. For the purpose of this study, we defined death as the moment that physical stimulation ceased to elicit muscle contractions. This state occurred within approximately 30 minutes of exposure to MgCl2. Specimens were not sterilized or chemically treated in any way prior to the experiment and were kept alive in continuously circulating seawater for any time spent in the laboratory prior to euthanasia. Microbes involved in post-mortem decay, therefore, would have consisted of those already found in the gut contents of the specimens as well as on the ectoderm.

The decay experiment methodology generally followed that of Sansom et al. 2010a on chordate decay, but with some modifications: after 30 minutes of exposure to the MgCl2 and artificial seawater mixture, specimens were placed in individual Petri dishes filled with the same solution. Specimens were placed into a trough made by bisecting a plastic straw down its length.

The purpose of the straw was twofold. First, it prevented movement of the worm post-mortem during transport into and out of the incubator. Second, it helped to keep the remains intact and generally contiguous during the decay process, thus mimicking the potential effects of being entombed in mud layers. To limit gas exchange and create conditions leading to dysoxia, Petri dishes were sealed with Parafilm immediately prior to incubation. Specimens were incubated at

17°C for the following number of days: S. pusillus, 10 days, H. planktophilus, 5 days, and B. occidentalis, 8 days.

Characters scored for their decay rates are described in Figure 2 and Table 1. Decay gradients were evaluated on a scale of 1 to 5, with the following definitions: (1) pristine characters showing no signs of decay, (2) superficial decay, such as wrinkling or fraying, (3) significant wrinkling and fraying and/or loss of structural rigidity, (4) advanced decay but with characters still recognizable, and (5) characters unrecognizable or completely disintegrated.

Observation intervals were chosen based on pre-experimental trials (every 12 hours for S. pusillus, every 6 hours for H. planktophilus, at 48 hours, 72 hours, 96 hours and 168 hours for B. occidentalis) which determined the total time required for a specimen to progress from pristine to completely disintegrated — up to 10 days in the case of S. pusillus — with the exception of the gill bars and nuchal skeleton, which remained intact for the duration of the experiment. At each interval, three S. pusillus, two H. planktophilus and two B. occidentalis were removed from the incubator and observed under a stereomicroscope. The outer morphology was observed first, before specimens were dissected for observation of the internal organs. Each morphological character of interest was scored according to the pre-determined decay gradient scale.

Specimens of S. tenuis from the Royal Ontario Museum and the Smithsonian National

Museum of Natural History were compared with the results of the experiments to describe their anatomy with reference to extant Enteropneust morphology and the observed sequence of character decay.

RESULTS

Figure 3 shows the sequence of character decay for S. pusillus, H. planktophilus and B.

7 occidentalis. The decay of enteropneust characters was broadly grouped into four stages (Fig. 4 and 5). The time it took to enter and pass through each stage varied according to species, but each stage was descriptive of the level of decay of each character of interest relative to other characters. Table 1 summarizes the sequence of decay for each character. Among all species, the gill pores and pre-oral ciliary organ were the first characters to decay. The longitudinal proboscis muscles showed signs of early decay (lack of clear definition of the concentric rings of muscle bands), but were recognizable in a highly decayed form until the late stages of the experiment.

Among all species, the gill bars and nuchal skeleton were the most decay resistant.

During stage 1, the specimen appeared nearly pristine externally. The ectoderm had begun to fray, but was generally contiguous and resistant to tearing. The gill pores were not identifiable by the end of stage 1, but most internal organs remained in a pristine state of preservation comparable to those of freshly euthanized specimens. The most significant damage was generally in the posterior or intestinal trunk, i.e., posterior to the pharyngeal trunk. Ruptures may form in the trunk ectoderm, although the location of any given rupture along the trunk was variable. Gut contents were sometimes seen spilling into the surrounding environment, indicating that, in some specimens, the intestinal endoderm had also ruptured.

In stage 2, the specimen was intact but was clearly undergoing decay. The ectoderm had lost elasticity, and presented very little resistance to forceps. The proboscis coelom had collapsed and the longitudinal proboscis muscles had liquefied. These factors together presented a proboscis with little to no rigid structure. In the pharynx, the collagenous nuchal skeleton and gill bars remained undamaged and undisturbed topographically. The heart-kidney-stomochord complex had lost definition and was beginning to dissolve into the surrounding medium. In the trunk, damage that occurred during stage 1 continued to progress. Sections of the trunk may have

8 become detached from one another, or held together only along the ventral midline, where there is a thickening of the nerve plexus into a cord.

Stage 3 was characterized by the disintegration of all non-collagenous structures. The proboscis and general musculature of the trunk was completely without form. The straw maintained the overall shape of the enteropneust. The collagenous nuchal skeleton and gill bars were still contiguous and attached to each other by the basal lamina. The heart-kidney- stomochord complex was recognizable as a discoloured or bloody patch near the anterior tip of the nuchal skeleton.

Stage 4 was the final stage of decay, in which the entire specimen was an organic soup.

No internal structures maintain its shape. The nuchal skeleton and gill bars remained completely undamaged structurally, but were now freely disarticulated from what remained.

FOSSIL COMPARISONS

Proboscis.— There are two primary elements to consider when examining proboscis preservation in fossil enteropneusts in comparison with decay data: variation in general outline and preservation of internal organs.

Among S. tenuis and other tentatively described fossil enteropneusts, proboscis morphology, like in their extant counterparts, is generally oblong and oval-shaped. However, some examples show squarer, less oblong, or irregular outlines (Fig. 6A). These variations in shape could be due stress at the time of death and are reminiscent of what was observed in modern forms following euthanasia (Fig. 6B).

Potential internal features that have been identified within the proboscis include the heart- kidney-stomochord complex, the proboscis coelom, and the pre-oral ciliary organ (Caron et al.

2013). These characters, however, are highly prone to decay, particularly the pre-oral ciliary organ, which was consistently the second character to disappear, following the gill pores (Fig 3).

The proboscis coelom also rapidly collapsed and began to show significant decay early in the decay sequence. The heart-kidney-stomochord complex was more resilient and persisted into stage 3 decay, but only as a tissue discolouration near the anterior tip of the nuchal skeleton. It was unlikely to have been fossilized in its characteristic ovular or spade shape, except in the most exceptionally well-preserved specimens.

Given that three of the suggested internal structures of the proboscis have relatively low fossilization potential, we propose instead that the features recognized in S. tenuis are the longitudinal proboscis muscles. In many specimens, darker zones are present within the proboscis, and the extent and shape of these zones is variable across specimens. This could be related to the extent of decay of the longitudinal proboscis muscles, as shown in extant forms

(Fig. 6C and D).

Nuchal Skeleton and Gill Bars.— The nuchal skeleton and gill bars were the most decay- resistant features in the enteropneust anatomy (Fig. 3). Evidence of nuchal skeletons in S. tenuis include their placement and size relative to the rest of the body and wishbone-shaped morphology, which are consistent with the nuchal skeletons of extant enteropneusts (Fig. 7A, B compared with Fig. 2B). The skeletal arms are long and thin, pointing posteriorly towards the trunk, and join together to form a single, larger arm, pointing anteriorly.

Serial gill bars are the most common internal feature present in S. tenuis, and are often well preserved in comparison to the surrounding features. There was no evidence of structural damage to the gill bars in any of our decayed extant specimens. Likewise, no examples of S. tenuis present gill bars that appear to be fractured or otherwise damaged. The level of gill bar

10 articulation in S. tenuis indicates the extent of pre-fossilization decay of the pharyngeal area: most specimens feature highly detailed and well-articulated gill bars that have maintained their positions and remain in close association with neighbouring gill bars (Fig. 7A). Some are no longer parallel, that is, they exhibit a shifting of orientation relative to one another (Fig. 7B). The pharyngeal area of these specimens would therefore indicate that decay had progressed to stage

3, the first point at which the basal lamina had degraded sufficiently such that the gill bars moved independently.

Resistance of the Esophageal Organ.— In H. planktophilus, the esophageal organ is the most decay-resistant of the soft internal organs (Fig. 3B). By late stage 2 and early stage 3, the esophageal organ was often relatively undamaged, whereas most surrounding tissues and the digestive tube had liquefied or begun to disintegrate. In S. tenuis, visually distinct patches have been identified directly posterior and ventral to the last set of gill bars in the pharynx. This position, the characteristic oval shape of this darkened or reflective region, and the relative decay resistance of the esophageal organ make it a good candidate for preservation (Fig. 7C,D).

Outline and External Features of Dead Specimens.— Across all three enteropneust species, the gill pores were the first character to disappear. While it is not surprising that an exterior ectodermal feature did not last long in the decay process, it is significant because it implies that the absence of gill pores among fossil enteropneusts may be due to taphonomic effects and not necessarily a true phylogenetic absence. The clear visibility of internal features, such as the gill bars, also suggests the rapid decay of external tissues.

The genital wings of the ptychoderid B. occidentalis were also found to decay at a rate approximately equal to that of the rest of the trunk (Fig. 3B). This is to be expected, considering that they are extensions of the pharyngeal trunk.

The outlines of S. tenuis fossils, while clearly of vermiform shape, frequently exhibit signs of light decay, i.e., late stage 1 to stage 2. Moderately frayed or fuzzy borders are indicative of ectodermal decay, while more drastic and localized irregularities may be indicative of liquid seeping out of the body or fraying of the ectoderm.

The post-pharyngeal trunk often has the most irregular outline, consistent with both the naturally wrinkled shape of the trunk and the early onset of decay in the most posterior regions of the trunk (Fig. 3, 4 and 5).

DISCUSSION

Phylogeny.— A direct comparison of extant enteropneust morphology, the sequence of character decay, and fossil examples strongly support the description of S. tenuis as a harrimaniid-like enteropneust that constructed pterobranch-like tubes (Caron et al. 2013). Like harrimaniid worms, S. tenius has a nuchal skeleton, but lacks gill bar synapticles, genital wings, hepatic sacs, and elaborations of the collar.

The collagenous nuchal skeleton and gill bars are by far the most decay-resistant structures in enteropneust anatomy (Fig. 3). These structures are well defined in S. tenuis and, as such, are crucial when inferring the phylogeny of the species. Among extant groups of enteropneusts, the shape of the nuchal skeleton of S. tenuis is most like that of the harrimaniids.

It has thin skeletal cornua extending at least to the middle of the collar, in contrast to the thicker and blunter skeletal arms of ptychoderids and spengelids. The nuchal skeleton is nearly or entirely absent from the deep sea torquaratorids (Osborn et al. 2012).

The gill bars of S. tenuis are well defined and clearly lack any evidence of collagenous synapticles bridging between the primary and secondary gill bars. This represents a phylogenetic absence rather than a taphonomic loss because synapticules are materially identical to gill bars and nuchal skeleton, which are well preserved. The absence of gill bar synapticles is one of the features diagnostic of the family Harrimaniidae (and Torquaratoridae; Deland et al. 2010), lending further support to this phylogenetic association.

There is also no evidence of ptychoderid like hepatic sacs or genital wings in S. tenuis.

The genital wings and hepatic sacs in extant enteropneusts decay at the same rate as the ectoderm and other superficial features. We would therefore expect evidence of one or both of these structures among the many examples of S. tenuis with well-preserved external outlines. Genital wings have been found in the Jurassic enteropneust Mesobalanoglossus buergeri, further supporting their absence rather than a taphonomic loss in S. tenuis (Bechly and Frickhinger

1999).

An argument has been made for the association of S. tenuis with the deep-sea enteropneust clade Torquaratoridae (Halanych et al. 2013; Cannon et al. 2013) based solely on the resemblance between mucous tubes that are constructed by some torquaratorids, and that have been noted to persist after being evacuated by the enteropneust, and those of S. tenuis. This argument is weak because torquaratorids and S. tenuis are morphologically unalike, similarities between the mucous tubes and fossilized burrows are superficial, and torquaratorids live in the deep sea, unlike the relatively shallow water Burgess Shale, shelf fauna.

The most visually striking difference between the torquaratorids and S. tenuis is collar morphology. The unadorned collar region of S. tenuis stands in contrast to the more complex and often “appendaged” collar region of most torquaratorids. While we have not observed the decay

13 of any deep sea acorn worms, there is no reason to believe that their collar appendages differ significantly enough in composition from the collar itself to have caused their selective absence from the fossil record. Rather, hypothetical fossil torquaratorids should present some range of collar appendage preservation corresponding to the preservation level of other, more ubiquitous, enteropneust characters.

The nuchal skeleton of S. tenuis also distances it from the Torquaratoridae. The torquaratorids generally lack nuchal skeletons and, when they do possess them, they are reduced to a medial plate. The possibility that S. tenuis represents the direct ancestor leading to the torquaratorid line, and that the nuchal skeleton was subsequently lost, is also unlikely. This positioning would place S. tenuis in a close relationship to the ptychoderids, which we have already rejected based on ample morphological differences between S. tenuis and the

Ptychoderidae.

The visual similarity of the tubes is most likely misleading. In general, perhaps without exception, enteropneusts secrete copious amounts of mucus. The construction of mucous tubes has been observed in many enteropneusts and is not an exclusive feature of the torquarotorids

(Urata and Yamaguchi 2004; Gonzalez and Cameron 2009). Therefore, S. tenuis cannot be placed with the torquarotorids based exclusively on these secretions. Further, tube breakages, when present in S. tenuis, are typically abrupt terminations between the displaced fragments, indicating a rigid structure (Fig. 8), like the tubes of pterobranch hemichordates (Caron et al.

2013). The tubes constructed by torquaratorids appear to be mucous-based and have been inferred to contain proteinaceous elements that contribute to their longevity. Pliable tubes such as these seem more likely to tear and pull apart raggedly rather than break cleanly when exposed to

14 similar external forces, demonstrating a fundamental difference in composition and structure between the tubes constructed by S. tenuis and torquaratorids.

The decay rate of the torquaratorid tubes is also difficult to reconcile with the fossil record because they appear to have lost most of their obvious structure within 24 hours, and disintegrate between 48 and 60 hours post evacuation. This rapid decay is comparable to that of decay-prone organs, such as the gill pores and pre-oral ciliary organ, that disappear in stage 2

(and which we have found no conclusive evidence of in any specimen of S. tenuis). Unlike mucous secretions, the tubes of S. tenuis tubes consistently preserve high levels of detail, with relatively little decay (Fig. 8), further emphasizing a compositional difference between the mucous tubes of torquaratorids and the indeterminate fossil tubes. The rapid rate of mucous decomposition likely explains the total absence of fossilized mucous among Cambrian biota.

The combination of a rigid tube structure and the overwhelming similarity between S. tenuis and the harrimaniids suggests that tube dwelling is an ancestral trait for the hemichordates and that it was lost leading to the enteropneusts. S. tenuis represents a stem-enteropneust possessing the acorn worm apomorphic characters, gill bars and nuchal skeleton (hypothesis 3 in

Fig. 1).

Incongruities.— There are two primary incongruities between the decay data and fossils of S. tenuis. The first is the preservation of characters inconsistent with the stages of decay observed in our experiment (Fig. 9). Examples A and B in Figure 9 present partial gill bar disarticulation but relatively little decay of the proboscis and digestive trunk, i.e., the area posterior to the pharyngeal trunk. Based on the state of the gill bars alone, the fossils could be inferred to have reached late stage 3 to stage 4 prior to diagenesis, but based on the proboscis and post-pharyngeal trunk, the fossil appears to have begun preservation at stage 2. One possible

15 explanation is that, prior to or during a burial event, the pharyngeal areas of these specimens were damaged. Lesions breaking through the ectoderm could have accelerated the decay of the surrounding area, leading to early disassociation of the gill bars relative to the expected sequence of decay.

The second incongruity is the unusual positioning of otherwise well-preserved structures.

Some specimens show nuchal skeleton-shaped characters that are not perfectly congruous with their expected position. However, these structures are never far from the “neck” of the fossil, where the proboscis connects to the trunk and where the nuchal skeleton is located. These displacements take the form of slight shifts or rotations relative to their expected positions (Fig.

6A, B). These topographic differences can be explained by considering the sequence of enteropneust decay. At stage 4, the collagenous nuchal skeleton and gill bars are the only remaining undamaged characters. Soft tissue has lost all structural integrity and elasticity, and connective tissue, including muscle, has disintegrated or liquefied. By the end of this stage, the nuchal skeleton and gill bars are completely disarticulated from any connecting structures and can freely move through the enteropneust remains. It therefore seems likely that S. tenuis specimens displaying the aforementioned nuchal skeleton displacement, while still providing evidence of an undamaged nuchal skeleton shaped structure, died and proceeded to degrade to late stage 3, when they were then buried and subsequently fossilized. These individuals would represent the most advanced state of decay that could reliably produce fossils recognizable as S. tenuis. A more extreme example of this phenomenon is the 180º rotation of a nuchal skeleton- like structure in Figure 9C. There is no other evidence of significant post-mortem disturbance of this specimen, and therefore the drastic repositioning of the nuchal skeleton defies explanation based on decay data.

While significant inconsistencies between character preservation and the observed stages of decay are rare in S. tenuis, the Mazon Creek enteropneust taxon Mazoglossus ramsdelli differs fundamentally in preservation (Fig. 9D). M. ramsdelli clearly displays the tripartite enteropneust body plan of proboscis, collar, and trunk. Fossil outlines present varying degrees of clarity and regularity, representing varying degrees of pre-diagenetic decay. However, no specimens display any observable internal characteristics. This is a marked departure from the preservation style of

S. tenuis, of which there are abundant examples of gill bar and nuchal skeleton preservation, as well as consistent patterns of soft tissue preservation.

These inconsistencies between fossil presentation and the decay data do not invalidate the usefulness of decay experiments when considering taphonomy. Instead, they emphasize that, rather than being completely diagnostic of the processes that lead to fossilization, decay experiments conducted in non-actualistic conditions provide benchmarks against which fossil taxa description can be measured. Deviations from these benchmarks can themselves be informative as inferences can be made regarding the nature of fossilization conditions, which may differ from observed laboratory decay sequences.

Burgess Shale Taphonomy.— Burgess Shale type preservation begins with the rapid entombment of organisms by mud-silt flows (Allison and Brett 1995; Gabbott et al. 2008). The speed of these depositional events, in combination with the precipitation of carbonate cements acting as a permeability barrier at the sediment-water interface, quickly limited the availability of oxidants required for microbial decay, and thus impeded decomposition (Gaines et al. 2012).

Most preserved features remain as flattened (often 2-dimensional) carbonaceous compressions, although preservation of some features may have involved pyritization and phosphatization

(Butterfield 2002; Gabbott et al. 2004; Gaines et al. 2012).

While there is some debate regarding the extent to which Burgess Shale organisms were transported prior to fossilization (Gabbott et al. 2008), biostratinomic evidence shows that the amount of transport must have been limited and most fossils were preserved close to their living habitat (Caron and Jackson 2006).

The goal of our experiment was to quantify the rate of morphological decay in enteropneusts regardless of the peculiar taphonomic conditions that might have been present at the Burgess Shale. To this end, we did not investigate the role of water chemistry or sediment type (ie. pore size, chemical composition), nor attempt to describe the chemical pathways that may have lead to exceptional preservation. However, our observations of the unimpeded decay sequence and rate of morphological characters of the Enteropneusta provide the framework for interpreting the fossil anatomy of S. tenuis, future hemichordate fossils from other fossil deposits as well as make broader inferences regarding the potential paleo-environmental conditions of the

Cambrian.

Comparing the quality of preservation of S. tenuis with the sequence and rate of decay of extant enteropneusts can inform estimates of the speed of fossilization in the Burgess Shale.

Enteropneusts are entirely soft bodied organisms, and can progress from death to almost complete disintegration within the span of 5 to 8 days at 17 ºC. This short timeframe and the high quality of soft tissue preservation displayed by many Burgess Shale fossils suggests that mineralization must have occurred soon after death, and/or decay must have been halted or significantly slowed. The rapid decay rates of enteropneust tissues in this study therefore lends support to previous work that showed that Burgess Shale-type preservation is most likely the result of suppression of natural decay processes (Gaines et al. 2008; Gaines et al. 2012).

Another feature of taphonomy revealed by the speed and sequence of enteropneust decay is the extent of pre- versus post-burial decay present in S. tenuis specimens. Very few specimens are likely to have undergone significant pre-burial decay, because after stage 2, any directional, external force would have been capable of radically shifting the topography of the remains, and liquefying tissues would have leaked out and away from the body. The well-articulated nature of most of these fossils indicates that organisms were likely buried before significant decay had occurred, because S. tenuis corpses are held together in a manner similar to that seen in the extant enteropneusts held in place in our experiments using a plastic trough.

Our observations also support the theory that most specimens presenting high quality preservation likely did not undergo significant transport and experienced minimal disturbance post-burial (Caron and Jackson 2006). Enteropneust anatomy is extremely fragile; it is not uncommon for them to suffer trunk damage or tearing of the tissues surrounding the nuchal skeleton during extraction and transport. Similarly, post-burial disturbance would likely result in significant shifts in topology, contingent upon the level of decay at the point of disturbance.

There is, however, some evidence for movement when considering that S. tenuis specimens are found both as isolated fossils and in assemblages. Isolated fossils exhibit a much higher degree of resolution of fine details and tend to be more articulated specimens than specimens preserved in assemblages. In contrast, assemblages of S. tenuis often display particularly poor resolution; large collections of bodies crisscross and on top of each other (Fig.

10 left). The scattered orientation of the various specimens may be indicative of tumbling during a depositional event. In these cases of significant transport, there is greater evidence of damage to specimens, including detachment of the proboscis and possible tearing of muscle tissues (Fig.

10 right).

Finally, an important point is that we did not see any trend in morphological character loss consistent with the concept of "stem ward slippage." Stem-wards slippage is the proposed underlying bias toward erroneously interpreting fossil taxa as basal based on the preferential loss of synapomorphic characters before plesiomorphic characters during decay (Sansom et al.

2010a). Among enteropneusts, stem ward slippage was not observed. Phylogenetically informative characters may be either decay-susceptible or decay-resistant. Derived family specific characters such as the genital wings and hepatic sacs are prone to decay, but no more so than the ectoderm and other external features of the class Enteropneusta. On the other hand, the nuchal skeleton, an autapomorphy of the Enteropneusta, and the gill bars, a plesiomorphic character of the deuterostomes, are the most decay-resistant structure in enteropneust anatomy.

Resistance to decay is more appropriately thought of as a consequence of the size and composition of a morphological feature.

CONCLUSIONS

While decay data do not completely explain all patterns of preservation observed in fossilized enteropneusts, we argue that the true strength of decay data is that a comprehensive understanding of decay allows us to produce a laboratory benchmark against which we can evaluate the degree to which decay may have proceeded in fossil forms before the process of preservation stopped it.

Our study of the decay of extant enteropneusts reveals that the sequence of morphological character decay is consistent across the three species studied; these species represent a disparate taxonomic sampling of the group. However, despite this consistency, we did not observe any meaningful bias against synapomorphic character loss above plesiomorphic

20 characters. We conclude, therefore, that the stem-wards slippage hypothesis is not affecting our description and phylogenetic placement of S. tenuis.

Using decay data as a benchmark when analyzing the Cambrian fossil enteropneust S. tenuis, we have confidently identified all major structural features and conclude that it is most like members of the family Harrimaniidae. This conclusion reinforces the view (Caron et al.

2013) that the morphologically simple harrimaniid body plan has been conserved over 500 million years with little observed change.

ACKNOWLEDGEMENTS

We thank Peter Fenton for assistance in the collections and the director and staff of the Bamfield

Marine Sciences Centre. We also thank Simon Darroch and James D. Schiffbauer for their constructive comments which substantially improved the manuscript. The decay study was supported by CB Cameron's NSERC Discovery Grant. K. Nanglu's doctoral research is supported by fellowships from the University of Toronto (Department of Ecology and

Evolutionary Biology) and J-B Caron's NSERC Discovery Grant (#341944). This is Royal

Ontario Museum Burgess Shale project number 61.

REFERENCES

Alessandrello, A., G. Bracchi, and B. Riou. 2004. Polychaete, sipunculan and enteropneust

worms from the lower Callovian (Middle Jurassic) of La Voulte-sur-Rhône (Ardèche,

France). Atti della Società italiana di Scienze naturali e del Museo Civico di Storia

naturale di Milano 32:3–14.

Allison, P. A. 1986. Soft-bodied animals in the fossil record: The role of decay in fragmentation

during transport. Geology 14:979-981.

— — —. 1988. The role of anoxia in the decay and mineralization of proteinaceous macro-

fossils. Paleobiology 14:139-154.

Arduini, P., G. Pinna, and G. Teruzzi. 1981. Megaderaion sinemuriense ng. n.sp, a new fossil

enteropneust of the Sinemurian of Osteno in Lombardy. Atti della Societa Italiana di

Scienze Naturali e dei Museo Civico di Storia Naturalie di Milano 122:104-108.

Bechly, G., and K. A. Frickhinger. 1999. The Fossils of Solnhofen 2. Goldschneck-Verlag,

Germany.

Briggs, D. E. G., and A. J. Kear. 1993a. Decay of Branchiostoma: implications for soft-tissue

preservation in conodonts and other primitive chordates. Lethaia 26:275-287.

— — —. 1993b. Decay and preservation of polychaetes – taphonomic thresholds in soft-bodied

organisms. Paleobiology 19: 107-135.

— — —. 1994. Decay and mineralization of shrimps. Palaios 9:431–456.

Briggs, D. E. G., A. J. Kear, M. Baas, J. W. Leeuw, and S. Rigby. 1995. Decay and composition

of the hemichordate Rhabdopleura: implications for the taphonomy of graptolites.

Lethaia 28:15-23.

Cameron, C. B. 2005. A phylogeny of the hemichordates based on morphological characters.

Canadian Journal of Zoology 83:196-215.

Cameron, C. B., B. J. Swalla, and J. R. Garey. 2000. Evolution of the chordate body plan: New

insights from phylogenetic analysis of deuterostome phyla. Proceedings of the National

Academy of Sciences (USA) 97:4469-4474.

Cannon, J. T., B. J. Swalla, and K. M. Halanych. 2013. Hemichordate molecular phylogeny

reveals a novel cold-water clade of Harrimaniid acorn worms. Biological Bulletin

225:194-204.

Caron, J. B., and D. A. Jackson. 2006. Taphonomy of the Greater Phyllopod Bed community,

Burgess Shale. Palaios 21:451-465.

Caron, J. B., S. Conway Morris, and C. B. Cameron. 2013. Tubicolous enteropneusts from the

Cambrian period. Nature 495:503-506.

Casenove, D., T. Oji, and T. Goto. 2011. Experimental taphonomy of benthic chaetognaths:

implications for the decay process of Paleozoic chaetognath fossils. Paleontological

Research 15:146-153.

Chen, J. Y., Dzik, J., Edgecombe, G. D., Ramsköld, L., and G. Q., Zhou. 1995. A possible Early

Cambrian chordate. Nature 377: 720-722.

Chen, A., and D. Huang. 2008. Gill rays of the primitive vertebrate Yunnanozoon from Early

Cambrian: a first record. Frontiers of Biology in China 3: 241-244.

Conway Morris, S. The Burgess Shale (Middle Cambrian) fauna. 1979. Annual Review of

Ecology, Evolution and Systematics 10:327-349.

— — —. 2009. The Burgess Shale animal Oesia is not a chaetognath: A reply to Szaniawski.

Acta Palaeontologica Polonica 54:175-179.

Conway Morris, S., and J. B. Caron. 2012. Pikaia gracilens Walcott, a stem-group chordate from

the Middle Cambrian of British Columbia. Biological Reviews 87:480-512.

Darroch, S. A. F., Laflamme, M., Schiffbauer, J.D., and D. E. G. Briggs. 2012. Experimental

formation of a microbial death mask. Palaios 27: 293-303.

Deland, C., C. B. Cameron, T. H. Bullock, K. P. Rao, and W. E. Ritter. 2010. A taxonomic

revision of the family Harrimaniidae (Hemichordata: Enteropneusta) with descriptions of

seven species from the Eastern Pacific. Zootaxa, 2408:1-30.

Donoghue, P. C. J., and M. Purnell. 2009. Distinguishing heat from light in debate over

controversial fossils. BioEssays 31: 178-189.

Gabbot, S. E., J. Zalasiewicz, and D. Collins. 2008. Sedimentation of the Phyllopod Bed within

the Cambrian Burgess Shale Formation of British Columbia. Journal of the Geologic

Society 165: 307-318.

Gaines, R. R., D. E. G. Briggs, and Z. Yuanlong. 2008. Cambrian Burgess Shale-type deposits

share a common mode of fossilization. Geology, 36:755-788.

Gaines, R. R., E. U. Hammarlund, X. Hou, C. Qi, S. E. Gabbott, Y. Zhao, J. Peng, and D. E.

Canfield. 2012. Mechanism for Burgess Shale-type preservation. PNAS 109:5180-5184.

Gonzalez, P., and C. B. Cameron. 2009. The gill slits and pre-oral ciliary organ of Protoglossus

(Hemichordata: Enteropneusta) are filter feeding structures. Biological Journal of the

Linnean Society 98:896-906.

Halanych, K. M., J. T. Cannon, A. R. Mahon, B. J. Swalla, and C. R. Smith. 2013. Modern

Antarctic acorn worms form tubes. Nature Communications 4.

Maletz, J. 2014. Hemichordata (Pterobranchia, Enteropneusta) and the fossil record.

Palaeogeography, Palaeclimatology, Palaeoecology 398:16-27.

Mitchell, C. E., J . M. Melchin, C. B. Cameron, and J. Maletz. 2013. Phylogeny of the tube-

building Hemichordata reveals that Rhabdopleura is an extant graptolite. Lethaia 46:34-

56.

Osborn, K. J., L. A. Kuhnz, I. G. Priede, M. Urata, A. V. Gebruk, and N. D. Holland. 2012.

Diversification of acorn worms (Hemichordata, Enteropneusta) revealed in the deep sea.

Proceedings of the Royal Society B: Biological Sciences 279:1646-1654.

Sansom, R. S., and M. A. Wills. 2013. Fossilization causes organisms to appear erroneously

primitive by distorting evolutionary trees. Scientific Reports 3:2545.

Sansom, R. S., S. E. Gabbott, and M. A. Purnell. 2010a. Non-random decay of chordate

characters causes bias in fossil interpretation. Nature 463:797-800.

— — —. 2010b. Decay of vertebrate characters in hagfish and lamprey (Cyclostomata) and the

implications for the vertebrate fossil record. Proceedings of the Royal Society B

278:1150-1157.

— — —. 2013. Atlas of vertebrate decay: a visual and taphonomic guide to fossil interpretation.

Palaeontology 56:457-474.

Shabica, C. W., and A. A. Hay. 1997. Richardson’s Guide to The Fossil Fauna of Mazon Creek.

Northeastern Illinois University, Chicago.

Shu, D., Zhang, X., and L. Chen. 1996. Reinterpretation of Yunnanozoon as the earliest known

hemichordate. Nature 380: 428-430.

Swalla, B. J., and A. B. Smith. 2008. Deciphering deuterostome phylogeny: molecular,

morphological and palaeontological perspectives. Philosophical Transactions of the

Royal Society B-Biological Sciences 363:1557-1568.

Szaniawski, H. 2005. Cambrian chaetognaths recognized in Burgess Shale fossils. Acta

Palaeontologica Polonica 50:1-8.

— — —. 2009. Fossil chaetognaths from the Burgess Shale: A reply to Conway Morris. Acta

Palaeontologica Polonica 54:361-364.

Urata, M., and M. Yamaguchi. 2004. The development of the enteropneust hemichordate

Balanoglossus misakiensis. Zoological Science 21:533-540.

Wilson, L. A., and N. J. Butterfield. 2013. Sediment effects on preservation of Burgess Shale-

type compression fossils. Palaios 29: 145-153.

Tables

Character Description Pattern of decay Proboscis The anterior soma or segment of the enteropneust  Consistently show early decay: fraying of the ectoderm and loss of that, when pulled up against the collar soma, often external definition. results in an acorn shape, hence the common name  Occasionally rupture, usually at anterior most point. ‘acorn worm.’  Total disintegration usually did not occur until late stage 3 due to some tissue remaining intact at the anterior nuchal skeletal region. Longitudinal proboscis Arranged diffusely in Harrimania, in concentric  Decay begins as loss of muscle definition. muscles (1) rings in Saccoglossus and as radial plates in  Could remain present at later stages of decay as undefined masses Balanoglossus (see Deland et al. 2010). of liquefied tissue. Proboscis coelomic cavity Fluid filled cavity of variable size that contains the  Early loss of fluid and deflation from a characteristic round (2) heart-kidney-stomochord complex. structure by the end of stage 1 to early stage 2. Heart-kidney-stomochord Complex of three structures: a contractile  Decay began with loss of external regularity as the pigmented complex (3) pericardium (heart), that pressurizes blood in the glomerulus liquefied. heart sinus against a turgid rod of cells  Decay proceeded inwards as the stomochord cell walls ruptured. (stomochord), resulting in the filtration of blood  At times persisting into stage 3 as a discoloured packet of tissue waste through the thin glomerulus (kidney). and blood. Pre-oral ciliary organ (4) A U-shaped band of cilia on the posterior  Decayed rapidly and was always absent by the end of stage 2. proboscis that transports particles into the mouth (Gonzalez and Cameron 2009). Nuchal skeleton (6) A ‘Y’ shaped collagenous structure that underlies  Remained pristine throughout the experiment, showing no signs of the proboscis stomochord anteriorly and bifurcates decay. posteriorly into the dorsal to dorsolateral collar  Connected to the gill bars by the basal lamina well into stage 3. coelomic cavities.  Detached from gill bars in stage 4. Gill pores (7) Paired, dorsolateral ectodermal pores that connect  Consistently the first characters to disappear with the breakdown of the gill slits and atrium to the external the surrounding ectoderm. environment.  Consistently absent by stage 2. Gill bars (8) Serially paired dorsolateral collagenous bars that  Remained pristine throughout the experiment, showing no signs of are inverted ‘W’ shape, and develop from the decay. pharynx ectoderm.  Adjacent gill bars became disarticulated in stage 4. Trunk The posterior vermiform soma or segment of the  Highly decay prone, but did not completely disintegrate until stage enteropneusts that includes the pharynx, 4. esophagus, and intestine.  Intestinal ectoderm ruptured early.  Continued to fray and tear, until entire sections of the trunk were no longer contiguous.  Pharyngeal trunk less prone to decay and held together by the dorsal midline, basal lamina and paired collagenous gill bars.

Trunk ventral midline (9) Thin band of ectoderm overlying a longitudinal  Rate of decay varied between individuals of S. pusillus and B. thickening of the ventral nerve plexus, or ventral occidentalis. trunk nerve cord. In S. pusillus and B. occidentalis  Often persisting into stage 3, however local regions of the ventral there are well developed paired bands of midline would in some specimens disintegrate by stage 2. longitudinal muscles paralleling either side of this  Thick nerve plexus often last structure physically linking two midline. sections of the digestive trunk. Genital wings (10) Of the three species examined, these paired dorso-  Decay at a rate similar to other large features such as the proboscis lateral extensions of the trunk ectoderm are unique and trunk. to B. occidentalis (a ptychoderid).  Fraying and loss of structural rigidity in stage 2.  Total loss usually occurred by the end of stage 3. Esophageal organ (11) Shaped like opposing pillows and positioned  Most decay resistant of the soft internal structures. between the pharynx and gut, it functions to  In H. planktophilus, remained identifiable until stage 3. squeeze excess water from the food cord. In H. planktophilus, it is a deep red-brown colour. Trunk dorsal midline (12) A thin band of ectoderm overlaying a longitudinal  Rapid decay, usually lost in stage 2. thickening of the dorsal nerve plexus or dorsal  In S. pusillus, the paired ridges that parallel each side of the dorsal nerve cord. midline were quickly lost. Hepatic sacs (13) Blind finger-like extensions, or caeca, of the gut  Decay at a rate similar to other large features such as the proboscis that project from the mid dorsal trunk of B. and trunk. occidentalis (a ptychoderid). They increase the  Lose most rigid definition and fray significantly by stage 2. digestive and absorptive area of the gut.  Total loss occurred by the end of stage 3. Digestive tube (14) The tubular endodermal gut, or intestine, begins  Rapid and early decay, lost by stage 3. after the esophageal organ and terminates at the  Ruptures were observed within 2 days, often accompanied by anus. ruptures of the overlying ectoderm, spilling gut contents into external medium.

Figures

FIGURE 2. A, Generalized anatomy of the Enteropneusta (lateral view) and sequence of decay of major features. The division of tripartite body plan (proboscis, collar, trunk) is highlighted.

FIGURE 3. Sequence and rate of decay for three enteropneust species. A, Saccoglossus pusillus.

= level 2, orange = level 3, red = level 4, white = level 5 (absent). Asterisk indicates earliest point of gill bar/nuchal skeleton disarticulation.

FIGURE 4. Sequence of decay of Saccoglossus pusillus. Rows from top to bottom: proboscis, pharyngeal area, trunk. Columns from left to right: pristine, decay stages 1-4. Scale bars in rows

A and C, 1 cm, and in B, 1 mm. Time elapsed since euthanasia in hours in the top right of each cell.

C, Patchy local variation of muscle decay within the proboscis, compared with the patchy patterns of darkness/reflectivity in fossil taxa (D); A–C, clockwise from top left: ROMIP62124,

USNM 202841, ROMIP62123, USNM 202797; D, left to right: USNM 202183, ROMIP63129.

FIGURE 7. A, B, Preservation of clearly harrimaniid type nuchal skeletons (1) and gill bars (2). Scale bars

=1mm. Orientation of preservation may result in variable presentations of the nuchal skeleton. A: the nuchal skeleton is preserved on a plane parallel to the fossil face (ROMIP63128). B: the nuchal skeleton is preserved at a slight angle to the fossil face (USNM 202469). C and D: Preservation of the esophageal organ. The presence of a dark/reflective oblong patch at the end of the pharyngeal trunk after the gill bars corresponds to the anatomical position of the esophageal organ. The esophageal organ is the most decay- resistant soft character. C, USNM 202472. D, USNM 509806.

Breakages terminate more abruptly than would be expected for a mucus-based structure.

FIGURE 9. Incongruities between decay data and the fossil record (clockwise from top left:

ROMIP 63124, USNM 202780, ROMIP 63125). A, B, High degrees of pharyngeal decay relative to the proboscis and digestive trunk, an inversion of what would be expected. C, A nuchal skeleton structure, rotated 180 from its expected orientation. There is no evidence that the proboscis has been disturbed; thus the external factors that resulted in this orientation are unclear. D, The Mazon Creek taxon Mazoglossus ramsdelli; although the tripartite body outline of an enteropneust is unequivocal, there is no discernible preservation of internal features.

FIGURE 10. Left, Assemblages of S. tenuis may be indicative of a depositional event gathering several specimens together (USNM 202603). Scale bar, 1 cm. Right, Such assemblages typically display signs of damage related to movement (USNM 202603). 1, Likely a detached proboscis lost during motion. 2, A comparatively well-preserved pharyngeal area displaying high gill bar clarity. 3, Significant post-pharyngeal damage to the trunk. 4, An unidentifiable, disarticulated tissue mass.

CHAPTER 2

Cambrian suspension-feeding tubicolous hemichordates

This paper was originally formatted for and published in the journal BMC Biology, and is reproduced here with the permission of that journal. Supplemental data are included as appendices.

ABSTRACT

Background: The combination of a meager fossil record of vermiform enteropneusts and their disparity with the tubicolous pterobranchs renders early hemichordate evolution conjectural. The middle Cambrian Oesia disjuncta from the Burgess Shale has been compared to annelids, tunicates and chaetognaths, but on the basis of abundant new material is now identified as a primitive hemichordate.

Results: Notable features include a facultative tubicolous habit, a posterior grasping structure and an extensive pharynx. These characters, along with the spirally arranged openings in the associated organic tube (previously assigned to the green alga Margaretia), confirm Oesia as a tiered suspension feeder.

Conclusions: Increasing predation pressure was probably one of the main causes of a transition to the infauna. In crown group enteropneusts this was accompanied by a loss of the tube and reduction in gill bars, with a corresponding shift to deposit feeding. The posterior grasping structure may represent an ancestral precursor to the pterobranch stolon, so facilitating their colonial lifestyle. The focus on suspension feeding as a primary mode of life amongst the basal hemichordates adds further evidence to the hypothesis that suspension feeding is the ancestral state for the major clade Deuterostomia.

Keywords: Enteropneusta, Hemichordata, Cambrian, Burgess Shale

BACKGROUND

Hemichordates are central to our understanding of deuterostome evolution. The two classes (tubicolous Pterobranchia and vermiform Enteropneusta) are monophyletic [1–3], but are morphologically disparate (however, see [4, 5] for an alternate viewpoint of Pterobranchia as sister to the family Harrimaniidae within a paraphyletic Enteropneusta). Accordingly they give only generalized clues as to both the anatomy and mode of life of the last common ancestor as well as its connections to the sister phylum Echinodermata (collectively Ambulacraria). The resistant tubaria of pterobranchs (notably the Paleozoic graptolites [6]) provide an adequate fossil record, but in contrast that of the enteropneusts is almost nonexistent [7–9]. One exception is a tubicolous taxon (Spartobranchus tenuis) from the middle Cambrian Burgess Shale [10]. This enteropneust is closely comparable to extant harrimaniids, although its organic tube finds no modern counterpart [11]. The coeval Oesia disjuncta Walcott [12] has been compared to groups as diverse as annelids [12], appendicularian tunicates [13] and chaetognaths [14, 15], thus remaining in phylogenetic limbo. The proposed chaetognath affinity was refuted by Conway

Morris [16] and a hemichordate affinity briefly suggested instead, albeit without detailed re- observation of original specimens or consideration of new material. On the basis of hundreds of specimens from the newly discovered Marble Canyon fossil locality (Kootenay National Park,

British Columbia) [17], we not only identify Oesia as a primitive enteropneust but also demonstrate that it constructed the perforated tube-like fossils previously assigned to Margaretia dorus and interpreted as thalli of a green alga similar to Caulerpa [18].

RESULTS

Oesia possesses the canonical enteropneust body plan of proboscis, collar and elongate trunk (Figs. 1, 2a, b) but is unusual in that posterior to the pharynx there is a bilobed structure, rather than a vermiform intestine. Body length averages 53 mm (n = 187, size range 2.4–120 mm), but the width seldom exceeds 10 mm. The proboscis is relatively elongate (ratio of length to width is 1.35 ± 0.58) and variable in shape (Figs. 2a, d, e, g, h; Additional files 1A, D–F, 2A,

F–I, 3C, 4). A conspicuous ovoid area at the medial base of the proboscis appears darker or more reflective than the surrounding area (Fig. 2a–c, f; Additional files 2F–I, 3C). This is interpreted as the heart-kidney-stomochord complex [11]. More irregular structures across the proboscis probably represent decayed musculature (Fig. 2c; Additional file 2F–I). The collar is rectangular, but with rounded edges (Fig. 2a–c, f, g; Additional files 1A, D–F, 2F–I, 3C, 5D, E, G). In proportion it is shorter than the proboscis (average ratio is 0.39 ± 1.12) but has an equivalent width (average proboscis to collar width is 1.08 ± 0.23 mm). At the posterior margin of the collar

(Fig. 2b, c, f; Additional file 2F–I), a dark or reflective band probably represents the circumcollar ridge, while a thin, longitudinal structure between the proboscis base and collar (Fig. 2d, e;

Additional file 5A, B, D–I) is interpreted as the nuchal skeleton. The pharyngeal region houses a series (about 3 bars/mm) of approximately U-shaped gill bars (Fig. 2g–j; Additional file 5A, C) but is remarkable in that it occupies approximately 80 % of the trunk length (Fig. 2a, g;

Additional files 2A, C, 5D, E, G). The posterior end of the trunk is bulbous (Fig. 2b, g;

Additional files 1D–E, 2H, 4, 5D, E, G) and sometimes terminates in a bilobed structure (Fig. 2a, f; Additional files 1B, C, 2A–D, F, H) that is usually wider than long (average width-to-length ratio is 1.48 ± 0.63).

The preservation of Oesia (n = 45 in Marble Canyon, n = 6 in Raymond Quarry) inside

Margaretia (now a junior synonym of Oesia disjuncta) suggests an original association (Fig. 3a;

Additional file 6). Only single worms are found within tubes, suggesting a solitary mode of life, although due to breakage during transport, it is conceivable that tubes may have been inhabited by more than one worm (Fig. 3b–d). Typically the tube is at least twice the width of the worm, suggesting the worm could move freely within its dwelling (Fig. 3e–j; Additional files

3B, 7, 8). Three-dimensional preservation of both sides of the tube (Fig. 4d, e) shows that the internal cavity of the tube was spacious, and that the tube was at least semi-rigid. The total length and extremities of the tubes are poorly known. This is because of either prior breakage or concealment (Fig. 4a), but at least one end (presumably the top of the tube) appears rounded and closed (Additional file 9B, C).

Tubes with irregular undulations and lacking the spiral pattern were previously interpreted as prostrate subterranean rhizomes (Fig. 4.2-3 [18]). While the reassignment from alga to organically produced tube invalidates this identification, it remains plausible that subterranean, lateral extensions of the tube could serve as an anchor. In any individual the width of the tube is usually consistent along the length, but otherwise it varies considerably

(4–20 mm). Occasionally a tube shows one (Fig. 3i; Additional files 7A, C, 8A–C, 9D–F) or, more rarely, two bifurcations (Fig. 4c). Each bifurcates at approximately the same angle and has the same width as the primary tube. The tube wall is perforated by spirally arranged pores (about

10 openings per revolution; Fig. 4a, b). In a single tube pore size varies. Some may be almost closed, but others have diameters equivalent to about a third of the tube width (Fig. 4a, b, d, e;

Additional file 9A). Pore shape varies from circular to oblong ellipse and rhombic. That these might simply be taphonomic variations is less likely given that the specimens are preserved

53 parallel to the bedding plane (Fig. 4a, b, g–e; Additional file 9A). The margins of the pores tend to be raised, imparting a semi-corrugated texture to the external surface of the tube (Fig. 4a, b;

Additional file 9A). The tube is composed of narrow fibres (about 7 μm) that are braided and/or overlain in bundles (Fig. 4f, g).

Margaretia dorus is unlike any known species of Paleozoic algae. In particular, the combination of a fibrous composition and elaborate pore architecture are inconsistent with an algal grade of organization, as are its biotic associations and size in relation to well established

Cambrian macroalgae [19]. This in turn argues against Oesia being an example of inquilinism.

While the dozens of co-occurrences of O. disjuncta and its tube strongly suggest an original association, the preservation of large numbers of isolated Oesia specimens on single bedding surfaces (Additional files 3, 4) at Marble Canyon also needs an explanation. One possibility is that the association was facultative and Oesia could alternate between a tubicolous and non- tubicolous existence. Alternatively the worm may have been forced to vacate the tube as an en masse evacuation prior to final burial. This may be related to both the high-energy burial events [17] and the resultant dysoxic conditions that such events create [20], although this hypothesis is weakened by the lack of obvious exit structures (i.e. there is no evidence the worms could enter or leave the tubes at either end).

In this context, fragmentation of the tubes and dispersal during transport is perhaps a more plausible explanation as to how the worms became isolated. This appears to be reasonable given the observation that although tubes with a length of up to 544 mm are known (Fig. 4c), tubes of comparable width can be not only significantly shorter (e.g. Figs. 3b, 4c), but sometimes are even smaller than the worms themselves. A related observation is that along the tube margins showing evidence for breakage, the bundles of fibres may exhibit a pattern of ’unbraiding.’ This

54 suggests that originally the tubes were vulnerable to damage (Fig. 3b). The second factor is that in at least some cases the tube evidently serves to conceal the worm. For a worm to be readily visible, the tube either needs to be prepared mechanically, split more or less along the axis or be sufficiently degraded so as to allow a view of the interior. Accordingly, tubes showing such evidence of degradation also contain worms in an evident state of decay (Fig. 3b–h). In such cases worms are poorly preserved and are effectively reduced to a narrow band of reflective carbon (Fig. 3k–m). Worms in such late stages of decay also show a tendency to bend at sharp angles into semi-discrete sections (Figs. 2g, 3e, f, l, m). This appearance may represent adjacent sets of gill bars maintaining their articulation through attachment to the collagenous basal lamina, but at points where this basal lamina has degraded, the more acutely angled bending occurs [11].

DISCUSSION

Establishing Oesia as an enteropneust that inhabited the tube previously identified as the alga

Margaretia has significant implications for the Cambrian paleogeography and paleoecology of this group. Until now, Oesia was one of the rarest of Burgess Shale taxa and was restricted to the

Walcott Quarry [21]. At the coeval Marble Canyon locality, however, it is amongst the five most abundant taxa [17] and occupied a key trophic position. In marked contrast, Margaretia is recorded from various Burgess Shale sites in Laurentia (including the Stephen Formation of

British Columbia and the Spence and Wheeler Shales of Utah [18] — Additional file 11: Table

S1), eastern Yunnan, China [22] and further afield in Siberia (originally referred to as

Aldanophyton [18]) [23]. This expanded distribution suggests that enteropneusts were a significant component of many Cambrian communities (Additional file 6, Additional file 11:

Table S1).

Oesia also throws important new light on the early evolution of the hemichordates.

Construction of a large, complex and presumably metabolically costly tube is consistent with a sessile lifestyle. Given that pores appear to be present on all sides, we suggest that the tube (and branches) stood vertically, with the basal region embedded in the substrate and the top presumably closed (Fig. 5a). The porosity of the tube would have prevented dysoxia and also allowed access to the water column for filter feeding. Given that the tubes could exceed 50 cm

(Fig. 4c), this suggests a tiering level at least equivalent to (if not above) the tallest sponges known from the Burgess Shale.

The strikingly extended pharynx and numerous gill bars that were employed in suspension feeding are functionally convergent with the hyperpharyngotremy seen in the tunicates [24], cephalochordates [25] and some Paleozoic jawless fish [26] (Fig. 5a). More generally, however, the pharyngeal arrangement seen in Oesia suggests that within the hemichordates as a whole suspension feeding was primitive. Whilst a few members of the basal harrimaniids facultatively filter interstitial pore water [27, 28], in extant enteropneusts the primary mode is deposit feeding, consistent with their mostly infaunal existence. Such a migration from an epifaunal existence may have been in response to increased predation pressure and as a consequence entailed significant anatomical changes. Notably in Oesia the post- pharyngeal trunk appears to lack the esophageal organ, which in extant forms serves to remove excess water from the food cord (and presumably performed the same function in

Spartobranchus tenuis where it is also present; a summary of the main differences between S. tenuis and O. disjuncta can be found in Additional file 10). So too in the more derived taxa the

56 hepatic caeca increase the absorptive area, presumably reflecting the increased demands of deposit feeding.

Oesia shares with the co-eval tubicolous S. tenuis [10] a bulbous posterior structure which may have acted as an anchor. In Oesia, however, the claw-like arrangement points to a more active role in attachment and release, perhaps as a consequence of its inhabiting a commodious tube. This interpretation draws potential comparisons to the juvenile post-anal tail of harrimaniid enteropneusts. This tail serves in ciliary locomotion and as an attachment device and may also be the homologue of the pterobranch stalk [27]. In this context, the specialized posterior structures seen in S. tenuis and O. disjuncta may actually be ancestral features. If so, these were ultimately lost in the crown group Enteropneusta, but in the Pterobranchia they helped to pave the way towards coloniality.

CONCLUSIONS

While too few morphological characters are available to permit a meaningful cladistic analysis, the unique combination of characters found in O. disjuncta encourages us to present a preliminary re-interpretation of early hemichordate evolution. First, a tubicolous, epifaunal and solitary habit are evidently primitive. The fibrous filaments of the Oesia tube have some resemblance to the fusellar fibres seen in graptolites such as the Cambrian Mastigograptus [29], as well as the comparable periderm of rhabdopleurid [30] and cephalodiscid pterobranchs [31].

An important inference is that Oesia (and Spartobranchus) possessed secretory glandular cells, presumably homologous with those located on the cephalic shield of the pterobranchs.

The apparent absence of fibres in the tubes of Spartobranchus suggests that their loss may have preceded the loss of the tube itself. Concurrent with a shift to a burrowing and deposit feeding existence, the crown group enteropneusts abandoned the construction of such tubes.

In contrast, the tubes of pterobranchs (and correspondingly the posterior stalk) were elaborated in parallel with their miniaturization and sessile coloniality (Fig. 5b). Crucially, the unique mix of pterobranch and acorn worm characteristics seen in Oesia suggests that an extensive pharynx and undifferentiated trunk are basal to the hemichordates, whereas Spartobranchus is more derived and is basal to the acorn worms [11]. Future discoveries of new Cambrian hemichordates will help elucidate the hypothesized transformation of the posterior structures into the pterobranch stolon and critically unveil the order of both trait acquisition and loss during the early diversification of this phylum.

Finally, the evidence that primitive enteropneusts were suspension feeders is congruent with the hypothesis that suspension feeding represents the primitive mode of life in deuterostomes [32] as a whole. In particular, it is notable that this lifestyle is seen in early stem- group echinoderms [33] and stem-group ambulacrarians [34], and is inferred in the ur- ambulacrarians [35] as well as the more problematic vetulicolians [36] and yunnanozoans [37].

METHODS

Sediment overlaying sections of some specimens was removed using a micro-engraving tool with a carbide bit. Specimens were observed using a stereomicroscope and photographed using different illuminations, using direct or cross-polarized light on dry or wet specimens.

Backscatter scanning electron images were obtained to visualize fine anatomical features.

Measurements of morphology were made using the program ImageJ. A list of specimens used in the analysis can be found in Additional file 11: Table S1 [38–47].

ACKNOWLEDGEMENTS

We thank P. Fenton, D.H. Erwin and M. Florence and B. Lieberman for collections assistance at the Royal Ontario Museum, Smithsonian Institution and the University of Kansas Natural

History Museum respectively. We also thank S. Loduca for helpful comments on the manuscript regarding algal affinities and locality information as well as two anonymous reviewers for their constructive comments. Material for this study was collected under several Parks Canada

Research and Collections permits. The 2014 field expedition at Marble Canyon was partially funded by a National Geographic Society research grant to J.-B. Caron. K. Nanglu’s doctoral research is supported by fellowships from the University of Toronto (Department of Ecology and

Evolutionary Biology) and J.-B. Caron’s NSERC Discovery Grant (number 341944). This is

Royal Ontario Museum Burgess Shale project number 63.

AUTHORS CONTRIBUTIONS

KN and JBC took photos of specimens. KN took measurements of all specimens. All authors made observations and wrote the manuscript. All authors read and approved the final manuscript.

REFERENCES

1. Cameron CB. A phylogeny of the hemichordates based on morphological characters. Can J

Zool. 2005; 83:196–215.

2. Peterson KJ, Su YH, Arnone MI, Swalla B, King BL. MicroRNAs support the monophyly of

enteropneust hemichordates. J Exp Zool Part B Mol Dev Evol. 2013; 320:368–74.

3. Cannon JT, Kocot KM, Waits DS, Weese DA, Swalla BJ, Santos SR, et al. Phylogenomic

resolution of the hemichordate and echinoderm clade. Curr Biol. 2014;1–6.

4. Halanych KM, Cannon JT, Mahon AR, Swalla BJ, Smith CR. Modern Antarctic acorn worms

form tubes. Nat Commun. 2013; 4:2738.

5. Stach T. Larval anatomy of the pterobranch Cephalodiscus gracilis supports secondarily

derived sessility concordant with molecular phylogenies. Naturwissenschaften. 2013;

100:1187–91.

6. Mitchell CE, Melchin MJ, Cameron CB, Maletz J. Phylogenetic analysis reveals that

Rhabdopleura is an extant graptolite. Lethaia. 2013; 46:34–56.

7. Arduini P, Pinna G, Terruzzi G. Megaderaion sinemuriense n.g. n.sp. a new fossil

enteropneust of the Sinemurian of Osteno in Lombardy. Atti Soc Ital Sci Nat Mus Civ

Sotria Nat Milano. 1981; 122:104.

8. Maletz J. Hemichordata (Pterobranchia, Enteropneusta) and the fossil record. Palaeogeogr

Palaeoclimatol Palaeoecol. 2014; 398:16–27.

9. Alessandrello A, Bracchi G, Riou B. Polychaete, sipunculan and enteropneust worms from the

lower Callovian (Middle Jurassic) of La Voulte-sur-Rhône (Ardèche, France). Atti Soc

Ital Sci Nat Mus Civ Sotria Nat Milano. 2004; 32:3–14.

10. Caron J-B, Morris SC, Cameron CB. Tubicolous enteropneusts from the Cambrian period.

Nature. 2013; 495:503–6.

11. Nanglu K, Caron J-B, Cameron CB. Using experimental decay of modern forms to

reconstruct the early evolution and morphology of fossil enteropneusts. Paleobiology.

2015; 41: 460–78.

12. Walcott CD. Cambrian geology and paleontology. Middle Cambrian annelids. Smithson

Misc Collect. 1911; 57:107–44.

13. Lohmann H. Oesia disjuncta Walcott, eine Appendicularie aus dem Kambrium. Mitteilungen

aus dem Zool Staatsinstitut Zool Museum Hamburg. 1920; 38: 69–75.

14. Szaniawski H. Cambrian chaetognaths recognized in Burgess Shale fossils. Acta Palaeontol

Pol. 2005; 50:1–8.

15. Szaniawski H. Fossil chaetognaths from the Burgess Shale: a reply to Conway Morris

(2009). Acta Palaeontol Pol. 2009; 54:361–4.

16. Conway Morris S. The Burgess Shale animal Oesia is not a chaetognath: a reply to

Szaniawski (2005). Acta Palaeontol Pol. 2009; 54:175–9.

17. Caron J-B, Gaines RR, Aria C, Mángano MG, Streng M. A new phyllopod bed-like

assemblage from the Burgess Shale of the Canadian Rockies. Nat Commun. 2014;

5:3210.

18. Conway Morris S, Robison RA. More soft-bodied animals and algae from the Middle

Cambrian of Utah and British Columbia. Univ Kansas Paleontol Contrib. 1988; 122:1–

48.

19. Loduca ST, Wu M, Zhao Y, Schiffbauer JD, Leroy M, O'Neil E. Seaweed through time: the

early Paleozoic. The Geological Society of America. 2015.

20. Gaines RR, Droser ML. The paleoredox setting of Burgess Shale-type deposits. Palaeogeogr

Palaeoclimatol Palaeoecol. 2010; 297:649–61.

21. Caron JB, Jackson DA. Paleoecology of the Greater Phyllopod Bed community, Burgess

Shale. Palaeogeogr Palaeoclimatol Palaeoecol. 2008; 258:222–56.

22. Hu SX, Zhu MY, Luo HL, Steiner M, Zhao FC, Li GX, Liu Q, Zhang ZF. The Guanshan

biota. Kunming, China: Yunnan Science and Technology Press; 2013.

23. Ivantsov AY, Zhuravlev AY, Leguta AV, Krassilov VA, Melnikova LM, Ushatinskaya GT.

Palaeoecology of the early Cambrian Sinsk biota from the Siberian platform. Palaeogeogr

Palaeoclimatol Palaeoecol. 2005; 220:69–88.

24. Chang P, Bone Q, Carre C. Tunicate feeding filters. J Mar Biol Assoc United Kingdom.

2003; 83:907–19.

25. Ulrik H, Svane I. Filter feeding in lancelets (Amphioxus), Branchiostoma lanceolatum.

Invertebr Biol. 2015; 118:423–32.

26. Janvier P, Lund R. Hardistiella montanensis n. gen. et sp. (Petromyzontida) from the Lower

Carboniferous of Montana, with remarks on the affinities of the lampreys. J Vertebr

Paleontol. 1983; 2:407–13.

27. Cameron CB. The anatomy, life habits, and later development of a new species of

enteropneust, Harrimania planktophilus (Hemichordata: Harrimaniidae) from Barkley

Sound. Biol Bull. 2002; 202:182–91.

28. Gonzalez P, Cameron CB. The gill slits and pre-oral ciliary organ of Protoglossus

(Hemichordata: Enteropneusta) are filter-feeding structures. Biol J Linn Soc. 2009;

98:898–906.

29. Bates D, Kozlowska A, Loydell D, Urbanek A, Wade S. Ultrastructural observations on

some dendroid and graptoloid graptolites and on Mastigograptus. Bull Geosciences.

2009;84:21–6.

30. Mierzejewski P, Kulicki C. Cortical fibrils and secondary deposits in periderm of the

hemichordate Rhabdopleura (Graptolithoidea). Acta Palaeont Pol. 2003; 48:99–111.

31. Gonzalez P, Cameron CB. Ultrastructure of the coenecium of Cephalodiscus (Hemichordata:

Pterobranchia). Can J Zool. 2012; 1269:1261–9.

32. Simakov O, Kawashima T, Marlétaz F, Jenkins J, Koyanagi R, Mitros T, et al. Hemichordate

genomes and deuterostome origins. Nature. 2015; 527:459–65.

33. Zamora S, Rahman IA. Deciphering the early evolution of echinoderms with Cambrian

fossils. Palaeontology. 2014; 57:1105–19.

34. Caron J-B, Conway Morris S, Shu D. Tentaculate fossils from the Cambrian of Canada

(British Columbia) and China (Yunnan) interpreted as primitive deuterostomes. PLoS

One. 2010; 5:e9586.

35. Smith AB. Deuterostomes in a twist: The origins of a radical new body plan. Evol Dev.

2008; 10:493–503.

36. Ou Q, Conway Morris S, Han J, Zhang Z, Liu J, Chen A, et al. Evidence for gill slits and a

pharynx in Cambrian vetulicolians: implications for the early evolution of deuterostomes.

BMC Biol. 2012; 10:81.

37. Chen A, Huang D. Gill rays of primitive vertebrate Yunnanozoon from Early Cambrian: A

first record. Front Biol China. 2008; 3:241–4.

38. O’Brien LJ, Caron J-B. Paleocommunity analysis of the Burgess Shale Tulip Beds, Mount

Stephen, British Columbia: comparison with the Walcott Quarry and implications for

community variation in the Burgess Shale. Paleobiology. 2016;42:27–53.

39. Johnston PA, Johnston KJ, Collom CJ, Powell WG, Pollock RJ. Palaeontology and

depositional environments of ancient brine seeps in the Middle Cambrian Burgess Shale

at The Monarch, British Columbia, Canada. Palaeogeogr Palaeoclimatol Palaeoecol.

2009; 277:86–105.

40. Caron J-B, Gaines RR, Mángano MG, Streng M, Daley A. A new Burgess Shale-type

assemblage from the “thin” Stephen Formation of the southern Canadian Rockies.

Geology. 2010; 9:811–4.

41. Walcott CD. Addenda to descriptions of Burgess Shale fossils. Smithson Misc Collect. 1931;

85:1–46.

42. Resser CE, Howell BF. Lower Cambrian Olenellus zone of the Appalachians. Geol Soc Am

Bull. 1938; 49:195–248.

43. Resser CE. Middle Cambrian fossils from Pend Oreille Lake, Idaho. Smithson Misc Collect.

1938; 97:1–12.

44. Waggoner B, Hagadorn JW. An unmineralized alga from the Lower Cambrian of California,

USA. Neues Jahrb Geol Paläontol. 2004; 231:67–83.

45. Kimmig J, Pratt BR. Soft-bodied biota from the middle Cambrian (Drumian) Rockslide

Formation, Mackenzie Mountains, Northwestern Canada. J Paleo. 2015;89:51–71.

46. Krishtofovich AN. Discovery of Lycopodiaceae in the Cambrian deposits of eastern Siberia.

Dokl Akad Nauk SSSR. 1953; 91:1377. 1379.

47. Hu SX, Zhu MY, Steiner M, Luo HL, Zhao FC, Liu Q. Biodiversity and taphonomy of the

early Cambrian Guanshan biota, eastern Yunnan. Sci China Earth Sci. 2010; 53:1765–73.

Figures

Fig. 1

Schematic anatomy of Oesia disjuncta. Co: collar, Cr: circum-collar ridge, Dg: digestive groove,

Pr: proboscis, Hks: heart-kidney-stomochord complex, Gb: gill bars, Gp: gill pores, Mo: mouth,

Po: pores, Ps: posterior structure, Tr: trunk, Tu: tube. Dashed lines indicate transverse cross sections

Fig. 2

General morphology of Oesia disjuncta from the Burgess Shale. (Specimens in d, e and jcome from the Walcott Quarry; all other specimens come from Marble Canyon). a Note bilobed posterior structure and extended pharyngeal area (ROMIP63737, part and counterpart are superimposed at the dashed line). b, c Tripartite body plan and internal organs in the proboscis

(ROMIP63711). d, e Large proboscis and possible nuchal skeleton (USNM 509815), see also

Additional file 5A–C. f Well-developed bilobed posterior structure (ROMIP63713). g–i Details of the pharyngeal area (h, partial counterpart of g, highlighted by vertical dashed line; i is close- up of framed area in g, ROMIP63710). j Left and right pairs of gill bars preserved in lateral view

(USNM 277844). Direct light images: a, b, h; polarized light images: c–g, j; SEM image: i. Co: collar, Cr: circum-collar ridge, Dg: digestive groove, Dm: dorsal midline, Gb: gill bars, Hks: heart-kidney-stomochord complex, Ll: lateral side left, Lr: lateral side right, Ns: nuchal skeleton,

Pr: proboscis, Ps: posterior structure, Tr: trunk. Scale bars: a = 10 mm, b–e = 1 mm, f– h = 5 mm, i = 500 μm, j = 2 mm

Fig. 3

Margaretia dorus tubes and associations with Oesia disjuncta from the Burgess Shale.

Specimens in a and d come from the Raymond Quarry; all other specimens come from Marble

Canyon. (a–h) Taphonomic gradient of the worm inside its tube from generally poorly preserved

(ROMIP63957). g Poorly preserved trunk and faded tube (ROMIP63952). h Close-up of framed area in g on counterpart, showing gill bars readily visible. i, jSpecimen showing clear tripartite body plan and evidence of gill bars (ROMIP63715). k The extant acorn worm Saccoglossus pusillus after 48 hours of decay at 25 °C showing dissociated parts, although the tripartite body plan is still recognizable. l, m O. disjunctaoutside of its tube, showing extreme signs of decay comparable with k. Direct light (l) is contrasted with polarized light (m) to reveal different aspects of fossil morphology (ROMIP63954). The ectoderm is fraying off, the proboscis is indistinct and the trunk has lost turgidity. Most worms preserved inside their tubes show a similar level of preservation. Direct light images: a, b, d, l; polarized light images: c, e–i, m. Bi: node of bifurcation, Fe: fibrous elements, Wo: worm, other acronyms see Figs. 1 and 2. Scale bars: a–c, e–g, k–m = 10 mm, d, i = 5 mm

Fig. 4 a, b Spirally arranged pores perforate the tube (ROMIP63716; see also Additional file 9A). cTwo examples of multiple bifurcation points in a single specimen. Extreme size variation underscores the fragmentary nature of most tubes (left: KUMIP 204373, right: KUMIP

147911). f, g Close-up of the pores and fibrous texture of the tube. Individual fibres are micrometre small (ROMIP63705). Bi: node of bifurcation, Fe: fibrous elements, Lo: lower surface, Po: pores, Up: upper surface, Wo: worm, other acronyms see Fig. 2. Scale bars: a, b, f, g = 5 mm, c–e = 10 mm

Fig. 5 a Life reconstruction with hypothetical closed terminal ends of the tubes — part of one tube partially removed to show a worm (drawing by Marianne Collins). b Phylogenetic relationship of

Deuterostomia derived from [2]. Mapping of characters based on [1, 2] with our proposed hypothetical position for Oesia disjuncta as a basal hemichordate (dashed line with question mark). The position of Spartobranchus tenuis is based on a taphonomic study of extant and fossil enteropneusts [11]. Character states: 1) pharyngeal gill bars, suspension feeding; 2) notochord; 3) tubicolous; 4) miniaturization, coloniality; 5) fuselli; 6) loss of tubicolous lifestyle, deposit feeding; 7) indirect development via tornaria larva; 8) stereom, water vascular system

Appendices

Appendix 1

High-resolution imagery of the original three Oesia disjuncta specimens figured by C.D.

Walcott, 1911, from the Burgess Shale (Walcott Quarry). (A–C) Lectotype (A, part; B, C, counterpart). The serial striations throughout the trunk initially lead Walcott to place this animal amongst the Annelida. They are re-interpreted as gill bars throughout an extended pharynx. This specimen also shows the posterior bilobed structure (USNM 57630). (D, E) The proboscis, collar, trunk and gill bars are all apparent (USNM 57631). (F) The proboscis, circum-collar ridge and gill bars are extremely pronounced (USNM 57632). Direct light images: A, C, D, F; polarized light images: B, E. For acronyms, see Fig. 1. Scale bars: 10 mm.

Appendix 2

Oesia disjuncta from the Burgess Shale (all specimens from Marble Canyon except USNM

203001, 203033 (C–E – Walcott Quarry). (A–B) Specimen showing the proboscis, gill bars, and exceptional preservation of the posterior bi-lobed structure (B = close-up of framed area in A)

(ROMIP63714). (C–E) Nearly complete specimen — anterior region (to the left) missing; pharyngeal gill bars extending almost completely to the terminal end of the trunk and bilobed posterior structure (D, E = close-ups of framed areas in C) (USNM 203001, 203033). (F–G)

Complete specimen showing rounded proboscis, collar, gill bars and bi-lobed posterior structure

(ROMIP63709). (H, I) Complete specimen showing rounded proboscis, kidney-heart- stomochord complex, collar, circum-collar ridge, gill bars and posterior structure. A patterning texture in the proboscis suggests possible proboscis muscles (ROMIP63707). Direct light images: A, F, H; Polarized light images: C, D, E, G, I. For acronyms see Fig. 1. Scale bars: A, C,

F, G, =5 mm; B, D, E=1mm; H=10 mm; I=2 mm.

Appendix 3

Cluster of Oesia disjuncta from the Burgess Shale (Marble Canyon) preserved on the surface of one large slab (ROMIP63735). O. disjuncta is highly gregarious at the Marble Canyon, occurring in high abundance across all stratigraphic levels (see also Additional file 7). (A) Overall slab.

(B–C) Close-ups of framed areas in A. The specimen on the left in B is preserved in a decayed tube. The specimen in C shows tripartite body plan, gill bars and kidney-heart-stomochord complex. Polarized light images: A–C. Acronyms see Fig. 1. Scale bars: A = 10 cm; B, C = 1 cm.

Appendix 4

Close up of a cluster of Oesia disjuncta from the Burgess Shale (Marble Canyon) preserved on the surface of one large slab (ROMIP63736). Specimens show clear tripartite body plan as well as variation in proboscis size and shape. Direct light image: A; Polarized light image: B. For acronyms, see Fig. 1. Scale bars = 1 cm.

Appendix 5

Previously un-figured Oesia disjuncta specimens from the Burgess Shale Walcott's Quarry –

Smithsonian Institution collection. (A–C): Partially dissociated specimen showing nuchal skeleton and gill bars (B and C = close ups of framed areas in A) (USNM 509815) – see also

Fig. 1d–e. (D–F) Complete specimen with proboscis slightly dissociated from the rest of the body. The thin element at the posterior of the proboscis is likely the nuchal skeleton. Gill bars are also visible, and the posterior structure is preserved laterally (F = close up of framed area in

E) (USNM 202440). (G–I) Complete specimen possibly showing the nuchal skeleton (USNM

202145, 203031). Direct light images: D, G; Polarized light images: A–C, E, F, H, I. For acronyms, see Fig. 1. Scale bars: A, D, E: 5 mm; B, C: 1 mm; F, I: 2.5 mm; G, H: 10 mm.

Appendix 6

Stratigraphic variation in abundance of Oesia disjuncta and Margaretia dorus in the Marble

Canyon paleocommunity. Bars and diamonds indicate numerical abundances of each taxon across 10 cm stratigraphic bins. Coloured bars indicate the percentage of the total number of specimens found within that bin from O. disjuncta and M. dorus (total community size estimates from 2012 and 2014 field collections). Numbers next to the bars indicate the number of occurrences of Oesia preserved inside of Margaretia observed within that bin. Stratigraphic levels on the vertical axis represent negative meters from the boundary between the Eldon and

Stephen Formations as a reference point.

Appendix 7

Detailed imagery of ROMIP63738 (see also Fig. 2d, e). Reflective areas in direct light tend to appear black using polarized light, emphasizing for example the kidney-heart-stomochord complex and the posterior structure (B, D are close-ups of framed areas in A and C). Direct light images: A, B; polarized light images: C, D. For acronyms see Fig. 1 and Fig. 2. Scale bars: A, C:

10 mm; B, D: 5 mm.

Appendix 8

Additional specimens of Oesia disjuncta preserved inside branching tubes from the Burgess

Shale (Marble Canyon). The worms show a typical high degree of reflectivity in contrast with the surrounding tubes. (A, B) ROMIP63712. (C–E) ROMIP63708. Direct light images: A, D; polarized light images: B, C, E. Br1: branch 1, Br2: branch 2, other acronyms see Fig. 1 and

Fig. 2. Scale bars: A, B, E: 5 mm, C, D = 10 mm.

Appendix 9

Morphology of Margaretia dorus from the Burgess Shale (Raymond Quarry: A, C, D; Marble

Canyon: E; Utah: B; Monarch Cirque: F). (A) Close-up of terminal end showing evidence of limited breakage and variability in pore sizes and shapes. The margins of the pores are upraised, giving the tube a semi-corrugated appearance when viewed laterally (ROMIP911390). (See also

Fig. 3a.) (B) Two specimens from the Spence Shale with possible rounded terminal ends

(see arrows) (ROMIP59635). (C, D) Specimen showing a central hole at one end, presumably representing the insertion of a more or less perpendicular branching tube (ROMIP63739). (E)

Specimen showing a circular structure corresponding to a node of bifurcation similar to

ROMIP63739 (ROMIP63706), with a possible rounded terminal end (see arrow). (F) Complete specimen also illustrated in Fig. 3g–h(framed area see Fig. 3h). This branching tube shows that pore density decreases near the node of bifurcation, that pore shape varies from rhomboid to more ellipse shaped, and that individual fibres are long and continuous through large sections of the tube (ROM 63716). Direct light images: A, B, E; polarized light images: C, D, F. Te: terminal end; for other acronyms, see Fig. 3. Scale bars: A, E: 5 mm, B, C, D, F: 10 mm.

Appendix 10

Major morphological differences between the two Cambrian, tubicolous enteropneusts Spartobranchus tenuis (top box) and Oesia disjuncta (middle box): (1) S. tenuis is thin and elongate and the trunk much more variable in width compared to O. disjuncta, which is stout and does not vary in width across the length of the trunk; (2) the pharyngeal gill bars are restricted to approximately 10–20 % of the total trunk length [10] in S. tenuis, but extend approximately 80 % of the total trunk length in O. disjuncta; (3) S. tenuis possesses an esophageal organ while O. disjuncta does not; (4) S. tenuis has a bulbous terminal structure while O. disjuncta has a claw-shaped terminal apparatus; also the tube of S. tenuis has an externally corrugated but smooth texture with no evidence of pores or openings (bottom box;

Species References Institution Countr Formation Localities (sub-localities) Specimens Figured in this study y observed* M. dorus Royal Ontario Museum (ROM) Canada "thick" Stephen Collins Quarry, Fossil Ridge, Yoho National Park, British Columbia 8 None Shale M. dorus Above Raymond Quarry (Tuzoia beds), Yoho National Park, British 3 None Columbia M. dorus Raymond Quarry, Fossil Ridge, Yoho National Park, British Columbia 24 Fig. 3A,D, Fig. 4A,B, Additional file 6A,C,D M. dorus Walcott Quarry, Fossil Ridge, Yoho National Park, British Columbia 2 None

M. dorus Talus material Fossil Ridge, Yoho National Park, British Columbia 124 None (ROM's AW, WT, UE) M. dorus Mount Field, Yoho National Park, British Columbia (ROM's AC, FC, FRS, 30 None SWF) M. dorus O'Brien and Caron 2015 Mount Stephen (Tulip Beds), Yoho National Park, British Columbia 2 None [38] M. dorus Mount Stephen, Yoho National Park, British Columbia (ROM's ESA, 41 None ESC, ESG1, ESG2, SS, ST, WS) M. dorus Odaray Mountain, Yoho National Park, British Columbia 3 None

M. dorus Curtis Peak, Yoho National Park, British Columbia 2 None

M. dorus Johnston et al. 2009 [39] Monarch Cirque, Kootenay National Park, British Columbia 3 Additional file 6F

M. dorus Caron et al. 2014 [17] Marble Canyon, Kootenay National Park 1191** Fig. 3B,C,E-J,L,M, Additional files 4, 5, 6E, M. dorus Caron et al. 2010 [40] "thin" Stephen Stanley Glacier, Kootenay National Park, British Columbia 9 None Shale M. dorus Chimney Peak, Kootenay National Park, British Columbia 2 None

M. dorus Johnston et al. 2009 [39] Duchesnay unit Miller Pass, Assiniboine Provincial Forest, British Columbia 2 None (Bolaspidella Zone) M. dorus USA Marjum House Range, Utah 5 None

M. dorus USA Wheeler Ampitheatre, House Range, Utah 5 None

M. dorus Walcott, 1931 [41] Smithsonian Institution (USNM) Canada "thick" Stephen Walcott or Raymond Quarry, Fossil Ridge Yoho National Park, British 71 None Shale Columbia M. dorus (stosei, ramosa) Resser & Howell 1938 [42] USA Kinzers Lancaster county, Pennsylvania 5 None

M. angustata Resser 1938 [43] USA Rennie Idaho 0 None

M. chamblessi Waggoner and Hagadorn, Los Angeles County Museum of USA Latham Marble Mountains, California 0 None 2004 [44] Natural History M. dorus Kimmig and Pratt 2015 [45] Royal Tyrell Museum Canada Rockslide Mackenzie Mountains, NWT 0 None

M. dorus Conway Morris & Kansas Museum of Invertebrate USA Langston (Spence Wellsville Mountains, Utah 3 Fig. 4C-E, Additional file Robinson, 1988 [18] Paleontology (KUMIP) Shale) 6B M. dorus Conway Morris & USA Wheeler Ampitheatre, House Range, Utah 0 None Robinson, 1989 [18] Margaretia Krishtofovich 1953 [46], Not available Russia Sinsk Yakutsk, right bank of the Lena River in the vicinity of the mouths of the 0 None (Aldanophyton Ivantsov et al. 2005 [23] Achchagyy-Tuoydakh and Ulakhan-Tuoydakh rivers antiquissimum) Margaretia Krishtofovich 1953 [46] Not available Russia Inikan Not available 0 None (Aldanophyton antiquissimum) Margaretia sp. Hu et al. 2010 [47], Hu et al. Yunnan Institute of Geological China Wulongqing Kunming-Wuding and Malong-Yiliang areas, Eastern Yunnan, China 0 None 2013 [22] Sciences

* 0=known occurences but no specimens where observed in this study

** including 1129 specimens not collected

Appendix 11

Biogeographical distribution of the fossil Margaretia and index of specimens used in this study.

Reference numbers can be found in the main text.

CHAPTER 3 A new Burgess Shale polychaete and the origin of the annelid head revisited

This chapter is formatted for and was originally published in the journal Current Biology, and is reproduced here with the permission of that journal. Supplemental data are included as appendices.

SUMMARY

Annelida is one of the most speciose (~17,000 species) and ecologically successful phyla.

Key to this success is their flexible body plan with metameric trunk segments and bipartite heads consisting of a prostomium bearing sensory structures and a peristomium containing the mouth.

The flexibility of this body plan has traditionally proven problematic for reconstructing the evolutionary relationships within the Annelida. While recent phylogenies have focused on resolving the interrelationships of the crown-group [1–3], many questions remain regarding the early evolution of the annelid bodyplan itself, including the origin of the head [4]. Here we describe an abundant and exceptionally well-preserved polychaete with traces of putative neural and vascular tissues for the first time in a fossilized annelid. Up to three centimeters in length,

Kootenayscolex barbarensis gen. et sp. nov., is described based on more than five hundred specimens from Marble Canyon [5], and several specimens from the original Burgess Shale site

(both in British Columbia, Canada). K. barbarensis possesses biramous parapodia along the trunk, bearing similar elongate and thin notochaetae and neurochaetae. A pair of large palps and one median antenna project from the anteriormost dorsal margin of the prostomium. The mouth- bearing peristomium bears neuropodial chaetae, a condition that is also inferred in Canadia and

Burgessochaeta from the Burgess Shale, suggesting a chaetigorous origin for the peristomial

98 portion of the head and a secondary loss of peristomial parapodia and chaetae in modern polychaetes.

KEYWORDS

Annelida, polychaete, Burgess Shale, Cambrian explosion, body plan, prostomium, peristomium.

Annelid head evolution, Marble Canyon

IN BRIEF

Fossilized annelids are very rare. Nanglu and Caron describe an abundant and exceptionally well preserved new species from the 508-million-year old Burgess Shale. The new species sheds light on the origin of the annelid head and suggests a segmental origin for the mouth-bearing segment.

HIGHLIGHTS

 An abundant Cambrian polychaete preserves exceptional morphological details.

 The new species possesses a median antenna and large palps on the prostomium.

 Neuropodial chaetae are present on the mouth-bearing peristomium.

 A chaetigerous origin for the peristomial portion of the annelid head is proposed.

RESULTS AND DISCUSSION a) Systematic palaeontology

Phylum: Annelida Lamarck, 1809

Genus: Kootenayscolex barbarensis gen. et sp. nov.

Etymology: Kootenay for Kootenay National Park in British Columbia, Canada, where the

Marble Canyon fossil locality is located, and scolex, the Greek word for “worm”; barbarensis from Barbara Polk Milstein, a long-time supporter of the Royal Ontario Museum and Burgess

Shale research.

Holotype: ROMIP64388 (Figure 1A; Figure S2B)

Referred material: Paratypes: ROMIP62972, ROMIP63099.1, ROMIP64389-64398 In addition, ca. 500 additional specimens at the Royal Ontario Museum. See Star Methods for locality information.

Preservation: Specimens display a wide range of burial angles (Figure 1) consistent with being engulfed in fast moving bottom mudflow deposits [5]. The chaetae are rarely preserved, possibly as a result of decay, angle of burial and the composition and thinness of the individual chaetae.

Dark patches occur within the base of the palps and parapodia (Figures 1B, C, 2A-H; S3C), often coalescing between adjacent parapodia, and between the frontalmost parapodia and the head. In the palps, such areas may correspond to coelomic cavities (see Diagnosis), whereas they seem topographically concordant with muscle tissues and associated coelomic cavities within the parapodia [6]. These structures are phosphatized (Figure 1C, S1) which could support a musculature origin [7], although muscle tissues themselves are not preserved.

100

Diagnosis: K. barbarensis ranges from 1 mm (Figures 1D, 1I, and 1J) to ca. 30 mm and possesses up to 25 chaetigers (Figure 1F). The general shape of the body is elongate and is widest at approximately the halfway point (Figure 1D, 1F, and 1H).

The anteriormost unit, representing the prostomium, is roughly trapezoidal in dorso- ventral aspect, and widest at the anterior end (e.g., Figures 1B, 1K and 2A-C). A pair of long and flexible appendages extends from the anterior dorsal edge of the prostomium (e.g., Figures 1A-

D, 1F-H, 1K, and 2A-H; Figures S1A, S2A-F, and S3A-F), with a more dorsally located shorter appendage between them (e.g. Figures 1A, 1K, and 2C-E). These are interpreted as paired palps and a median antenna, respectively. The palps can reach 1/3 of body length (e.g., Figure 1H), possess a thick base and taper distally to a fine point (Figure 1B-D, 1F-H, and 1K). When twisted, the palps appear to become slightly flattened distally (Figure 1N). A thin reflective band rich in carbon and partially phosphatized, roughly 1/3 the diameter of the palps, runs from their distal tips to the anterior of the prostomium (e.g., Figures 1B, 1C, and 2A-C; Figures S1D, S1G, and S2A). An even darker, thinner carbon-rich structure traces the same pathway within this channel (Figures 1C, 2A, and 2C; Figure S1D-F), consistent with the neural architecture of many extant polychaetes, with neural pathways converging on a dense central neuropil region in the prostomium [8–11]. The larger, lightly shaded pathway within the palps is consistent with a coelomic cavity [7].

The median antenna is roughly half the thickness of the base of a palp and appears more rigid than the palps, further supporting its identification as an antenna [12]. A dark internal structure similar to those found in the palps seems to have a common origin within the prostomium (Figure 2H). Posteriorly, a pair of structures sometimes preserved laterally at the base of the palps (Figures 2A-G) or directed more anteriorly, and with radiating chaetae

101 projecting anteriorly or laterally, are interpreted as peristomial parapodia bearing chaetae. Only a single set of chaetal bundles could be observed along the peristomial parapodia, suggesting a uniramous condition. The preserved chaetal bundles are hypothesized to represent neurochaeta.

The mouth is between those parapodia, as suggested by the anteriormost portion of the gut

(Figures 1B, 1C, and 2A-C, S1), and by darker, single axial structures corresponding to the bases of these parapodia (Figures 1B, 1C, and 2A-D, S1). Three-dimensional patches of variable sizes and shape along various sections of the gut, and well beyond the mouth, suggest active mud infill

(Figures 1B, and 1K; Figure S1D, S2A).

Parapodia are identical in morphology on all segments (e.g., Figures 1D-F) and are large relative to the body width, up to half the width of the chaetiger to which they are attached. All chaetae are of the simple capillary type and notochaetae and neurochaetae are of similar morphology and length. They are extremely fine in comparison with other Burgess Shale polychaetes, at approximately 10–28 μm wide (e.g., Figure 2I-N). The longest chaetae relative to body size are about 2 to 4 times as long as the width of associated chaetigers (Figure 2I-N). A maximum of 16 neurochaetae radiate as a tightly clustered fan at an angle roughly perpendicular to the antero-posterior axis of the animal (Figure 2I-N). A maximum of 12 notochaetae project posteriorly, and tend to be slightly more fanned out (Figure 2I, and 2L).

The pygidium is indistinct, and there is no evidence of pygidial cirri.

b) Ecology

The presence of a median antenna, and the elongate sensory palps both suggest an active lifestyle. Although K. barbarensis has long chaetae, they do not appear to have any

102 specializations for swimming, as has been suggested for the Cambrian taxon Canadia spinosa

[13]. Like the majority of Cambrian forms (except for C. spinosa and potential infaunal forms such as Peronochaeta dubia[13]), K. barbarensis was most likely an epibenthic deposit feeder

(supported by gut infills), using its notochaetae for defense [14,15]. The uniramous parapodia on the peristomium and their positioning lateral to the mouth, may have speculatively supported buccal musculature useful for expanding the mouth to engulf food particles.

We performed a phylogenetic analysis using a revised matrix from [3] (see part (c) as well as Star Methods for a description of character coding changes and the parameters of our phylogenetic analyses). K. barbarensis with the Cambrian taxa Burgessochaeta setigera,

Phragmochaeta canicularis, and C.spinosa, fall into a stem-group position to the crown group as in previous analyses [15,16], although in a polytomy (Figure 4A). A complementary parsimony analysis is less well resolved, retrieving the same taxa in a polytomy with other members of the crown-group (S4). These results suggest that Cambrian forms, which are anatomically simpler than most modern forms in terms of body regionalization and specialization of chaetae, share a number of morphological similarities. This simple body morphology is predicted by ancestral state reconstructions based on phylogenomic and transcriptomic datasets [1,2], but there is less confidence regarding the head and its attendant structures [4]. Structures such as the nuchal organs and eyes are expected in the ancestral annelid but not found in the Cambrian fossil record, although this lack of evidence may be taphonomic, due to their generally small size and presumably low preservation potential. Although a median antenna is generally considered to be

103 a derived character in annelids, it is present in both errant and sedentary clades (ie. Phyllodocida,

Eunicida, Amphinomida, Paraonidae and some Spionida [6]), where it follows similar innervation patterns [9]. While possibly homologous, K. barbarensis is the only Cambrian polychaete with this structure suggesting an autapomorphy.

Parry et al. [15] recently proposed a “metameric hypothesis” for the origin of the annelid head, based on their reinterpretations of several Cambrian polychaetes (see also [3]). In their framework, the ancestral annelid would have appeared very similar to P. canicularis, that is, a uniformly bristled epibenthic worm with an anterior segment with biramous chaetae-bearing parapodia in lieu of a true prostomium. C. spinosa was suggested to represent a transitional form where the prostomial notopodia had developed into sensory palps while retaining prostomial neuropodia and neurochaetae.

The metameric head hypothesis is problematic for several reasons. One, it would imply that the last common ancestor of annelids and its closest sister group might bear a body completely covered in chaetae or sclerites, a position potentially occupied by wiwaxiids [17].

However the presence of an unambiguous molluscan radula in wiwaxiids [18] makes this scenario rather conjectural. The phylum level inter-relationships of the lophotrochozoans are also far from resolved, with molecular datasets routinely recovering either Nemertea or Brachiopoda as closer to Annelida than Mollusca [19,20], further calling into question the utility of the morphology of wiwaxiids for commenting on the anatomy of the ancestral annelid. While the most recent total evidence analysis recovers Wiwaxia corrugata, Kimberella quadrata and

104

Odontogriphus omalus as basal stem group molluscs [16], the significant morphological gap between these three taxa and the Cambrian polychaetes (with unambiguous parapodia-bearing chaeta) is also difficult to reconcile with an entirely chaetigerous ur-annelid (see also [12]).

The metameric head hypothesis [15] is also problematic in that the prostomium is unequivocally considered pre-segmental. In crown-group Annelida, the prostomium is derived from the region of the annelid embryo corresponding to the prototroch and above, while the segmental body is derived from the growth zone below the metatroch; this embryological division is among the most highly conserved patterns in annelid ontogeny [8, 16]. The metameric head hypothesis would require the de novo transformation of a segmental unit into a prostomial unit, and protostomial notopodia into sensory palps. While it is possible that such transitions occurred within the annelid stem lineage, the metameric head hypothesis does not posit any plausible mechanism for this alteration to the fundamental ground pattern of the crown-group

Annelida.

Considering P. canicularis to be close to an ancestral form, a central pillar of the metameric head hypothesis, is also problematic on taphonomic grounds, as this taxon is rare and poorly preserved [22]. Similar “head” morphologies, with bundles of chaetae pointing anteriorly, are found in many Burgess Shale polychaetes, including in B. setigera [20, 21] because the head is usually buried at an angle relative to the rest of the body in most specimens (Figure S3K).

Furthermore, when the metameric head hypothesis was developed, it relied on C. spinosa bearing a transitional prostomial morphology to illustrate the plausibility of a prostomium simultaneously bearing palps and notopodia+chaetae. While a complete redescription of C. spinosa is beyond the scope of this paper, we interpret the prostomium with ventral parapodia [5] as a parapodia-bearing segment directly posterior to a small prostomium, since there is no

105 evidence that the head is buried at an angle or deformed (S3G, H). A similar configuration is also seen in B. setigera (S3I, J). Like C. spinosa and K. barbarensis, the fact that this segment bears both the mouth and parapodia (presumably uniramous) + chaetae leads us to conclude that it is homologous with the peristomium of extant annelids.

(d) The chaetigerous mouth hypothesis

K. barbarensis, C. spinosa, and B.setigerea fall into a stem polytomy (Fig 4A) suggesting that their unique peristomial morphology is plesiomorphic and does not represent an autopomorphy of these Cambrian polychaetes. This leads us to propose a new hypothesis for the development of the modern annelid head. This is particularly true with regards to the location of the mouth, which up to this point has remained unclear in Cambrian annelids [15].

Rather than a metameric origin for the entire annelid head, whereby an anterior-most chaetigerous unit transitions into both prostomium and peristomium in the crown group (within which both are apodous+achaetigerous), we propose a segmental or chaetigerous origin for only the peristomial portion of the annelid head. While further fossil evidence is required to corroborate this view, our hypothesis suggests a total group annelid ancestor that is less of a radical departure from the ground pattern of the crown-group Annelida compared to Parry et al.

[3].

Using this framework, the ancestral annelid would have had a small prostomium with two palps (as is generally predicted [1,2,4]), and a segmented body, bearing a mouth on the first chaetiger, as well as biramous parapodia+chaetae (although as of yet there are no unequivocal fossils showing this biramous condition; Figure 4B-C). The uniramous peristomium found in K.

106 barbarensis, B. setigera, and C. spinosa, as well as the modern condition of an apodous peristomium, could have come about through one of two related developmental pathways.

The parapodia and chaetae of the ancestral mouth-bearing segment may have been lost in a similar manner as occurs during the development of at least three species within the genus

Magelona, which possess “transient parapodia” and “transient chaetae” [25]. These transient structures occur on segments 1 and 2 during the larval stage, but are subsequently lost after metamorphosis. The Cambrian uniramous peristomium condition may represent a “transitional state” whereby the notopodia were lost in the ancestrally biramous mouth-bearing segment through a similar developmental mechanism as in the magelonids (“transitional” stage in Figure

4B). The neuropodia and neurochaetae of this transitional peristomium would be lost before the advent of the crown group annelids, leading to the modern apodous and achaetigerous head.

Alternatively, Cambrian polychaetes may evidence a developmental process unique to the Cambrian, but which may also have parallels in the developmental biology of extant polychaetes, such as the nereidids. In the nereidid Hediste diversicolor [26] (also the genera

Laeonereis [27] and Platynereis [28]), the adult peristomium is developed through the fusion of the first larval chaetiger, which bears parapodia and chaetae, to the larval peristomium. The parapodia of the first larval chaetiger are re-absorbed or “aborted” during fusion, after which they form the posterior tentacular cirri. The uniramous mouth segment of K. barbarensis may have been derived from a similar process (Figure 4C), whereby: (1) the larvae of the Cambrian taxa may have appeared morphologically modern with an apodous mouth-bearing segment, (2) the first body segment fused with the mouth segment during ontogeny, and, (3) the notopodia and notochaetae were lost prior to adulthood using a developmental mechanism similar to the one seen in the nereiidids to produce the morphology seen in K. barbarensis, B. setigera and C.

107 spinosa. In this context, the modern polychaete head morphology was derived from a larva with an apodous peristomium (4), or a fused mouth and body segment may have subsequently lost the neuropodia and chaetae to give rise to the extant peristomial apodous+achaetigerous condition

(5).

An elevated degree of plasticity in gene regulatory networks during the dawn of the metazoans has been invoked as a contributor to the rapid appearance of nearly all major phyla during the Cambrian Explosion [28, 29]. Therefore, while direct evidence of developmental biology is difficult to document from early Palaeozoic fossils—although some data is available in fossilized embryos [31,32]—it should be expected that continued observations of early fossil taxa will yield morphologies that would have been difficult, if not impossible, to anticipate based solely on extant lineages. In this way, K. barbarensis joins similar discoveries of early arthropods [33], mollusks [16] and hemichordates [34] that complement ontogenetic studies to unveil the origins of body plans before developmental canalization concomitant with the establishment of modern phyla [35].

108

ACKNOWLEDGEMENTS

We thank P. Fenton and M. Akrami for collections assistance at the Royal Ontario Museum,

Sharon Lackie for elemental maps (Figure 1C and Figure S1B-G) and Danielle Dufault for illustrations (Figure 3). We thank D. de Carle for feedback and assistance with phylogenetic analyses. We also thank L. A. Parry and one anonymous reviewer for their comments which substantially improved the manuscript. Material for this study was collected under several Parks

Canada Research and Collections permits (to J.-B. Caron). Major funding support for field work comes from the Royal Ontario Museum (Research and collection grants), the National

Geographic Society (2014 research grant to J.-B. Caron), the National Science Foundation (2016

EAR-1556226 Award to Robert Gaines-Pomona College). K. Nanglu’s doctoral research is supported by fellowships from the University of Toronto (Department of Ecology and

Evolutionary Biology) and J.-B. Caron’s NSERC Discovery Grant (number 341944). This is

Royal Ontario Museum Burgess Shale project number 73.

AUTHOR CONTRIBUTIONS

Both authors conceived the project, made observations, analyzed the data, created figures and wrote the manuscript. K.N. took morphometric measurements. J.-B.C. led fieldwork activities and prepared and photographed the material.

DECLARATION OF INTERESTS

The authors declare no conflict of interest.

109

REFERENCES

1. Struck, T.H., Paul, C., Hill, N., Hartmann, S., Hösel, C., Kube, M., Lieb, B., Meyer, A.,

Tiedemann, R., Purschke, G., et al. (2011). Phylogenomic analyses unravel annelid

evolution. Nature 471, 95–98.

2. Weigert, A., Helm, C., Meyer, M., Nickel, B., Arendt, D., Hausdorf, B., Santos, S.R.,

Halanych, K.M., Purschke, G., Bleidorn, C., et al. (2014). Illuminating the base of the

Annelid tree using transcriptomics. Mol. Biol. Evol. 31, 1391–1401.

3. Parry, L.A., Edgecombe, G.D., Eibye-Jacobsen, D., and Vinther, J. (2016). The impact of

fossil data on annelid phylogeny inferred from discrete morphological characters. Proc. R.

Soc. B Biol. Sci. 283, 20161378.

4. Eibye-Jacobsen, D., and Vinther, J. (2012). Reconstructing the ancestral annelid. J. Zool.

Syst. Evol. Res. 50, 85–87.

5. Caron, J.-B., Gaines, R.R., Aria, C., Mángano, M.G., and Streng, M. (2014). A new

phyllopod bed-like assemblage from the Burgess Shale of the Canadian Rockies. Nat.

Commun. 5, 3210.

6. Rouse, G.W., and Pleijel, F. (2001). Polychaetes (Oxford: Oxford University Press).

7. Wilson, P., Parry, L.A., Vinther, J., and Edgecombe, G.D. (2016). Unveiling biases in

soft-tissue phosphatization: extensive preservation of musculature in the Cretaceous

(Cenomanian) polychaete Rollinschaeta myoplena (Annelida: Amphinomidae).

Palaeontology 59, 463–479.

8. Heuer, C.M., and Loesel, R. (2008). Immunofluorescence analysis of the internal brain

anatomy of Nereis diversicolor (Polychaeta, Annelida). Cell Tissue Res. 331, 713–724.

9. Orrhage, L., and Müller, M.C.M. (2005). Morphology of the nervous system of

110

Polychaeta (Annelida). Hydrobiologia 535/536, 79–111.

10. Zanol, J. (2010). Homology of prostomial and pharyngeal structures in Eunicida

(Annelida) based on innervation and morphological similarities. J. Morphol. 1043, 1023–

1043.

11. Heuer, C.M., Müller, C.H., Todt, C., and Loesel, R. (2010). Comparative neuroanatomy

suggests repeated reduction of neuroarchitectural complexity in Annelida. Front. Zool. 7,

13.

12. Eibye-Jacobsen, D. (2004). A reevaluation of Wiwaxia and the polychaetes of the Burgess

Shale. Lethaia 37, 317–335.

13. Conway Morris, S. (1979). Middle Cambrian polychaetes from the Burgess Shale of

British Columbia. Philos. Trans. R. Soc. B Biol. Sci. 285, 227–274.

14. Westheide, W. (1997). The direction of evolution within the Polychaeta. J. Nat. Hist., 1–

15.

15. Parry, L., Vinther, J., and Edgecombe, G.D. (2015). Cambrian stem-group annelids and a

metameric origin of the annelid head. Biol. Lett. 11, 1–5.

16. Vinther, J., Parry, L., Briggs, D.E.G., and Van Roy, P. (2017). Ancestral morphology of

crown-group molluscs revealed by a new Ordovician stem aculiferan. Nature 542, 471–

474.

17. Zhang, Z., Smith, M.R., and Shu, D. (2015). New reconstruction of the Wiwaxia

scleritome, with data from Chengjiang juveniles. Sci. Rep. 5, 1–9.

18. Smith, M.R. (2012). Mouthparts of the Burgess Shale fossils Odontogriphus and Wiwaxia:

implications for the ancestral molluscan radula. Proc. R. Soc. B Biol. Sci. 279, 4287–

4295.

111

19. Dunn, C.W., Hejnol, A., Matus, D.Q., Pang, K., Browne, W.E., Smith, S.A., Seaver, E.,

Rouse, G.W., Obst, M., Edgecombe, G.D., et al. (2008). Broad phylogenomic sampling

improves resolution of the animal tree of life.pdf. Nature 452, 745–749.

20. Kocot, K., Struck, T.H., Merkel, J., Waits, D.S., Todt, C., Brannock, P.M., Weese, D.A.,

Cannon, J., Moroz, L.L., Lieb, B., et al. (2017). Phylogenomics of lophotrochozoa with

consideration of systematic error. Syst. Biol. 66, 256–282.

21. Schroeder, P.C., and Hermans, C.O. (1975). Annelida: Polychaeta. In Reproduction of

Reproduction of Marine Invertebrates III: Annelids and Echiurans. A. C. Giese and J. S.

Pearse, eds. (New York: Academic Press).

22. Conway Morris, S., and Peel, J.S. (2008). The earliest annelids: lower Cambrian

polychaetes from the Sirius Passet Lagerstätte, Peary Land, North Greenland. Acta

Palaeontol. Pol. 53, 137–148.

23. Caron, J.-B., and Jackson, D.A. (2006). Taphonomy of the Greater Phyllopod Bed

Community, Burgess Shale. Palaios 21, 451–465.

24. Caron, J.-B., and Jackson, D.A. (2008). Paleoecology of the Greater Phyllopod Bed

community, Burgess Shale. Palaeogeogr. Palaeoclimatol. Palaeoecol. 258, 222–256.

25. Wilson, D.P. (1982). The larval development of three species of Magelona (Polychaeta)

from localities near Plymouth. J. Mar. Biol. Assoc. United Kingdom, 385–401.

26. Dales, R.P. (1950). The reproduction and larval development of Nereis diversicolor O. F.

Muller. J. Mar. Biol. Assoc. UK 29, 321–360.

27. Mazurkiewicz, M. (1975). Larval development and habits of Laeonereis culveri (Webster)

(Polychaeta: Nereidae). Biol. Bull., 186–204.

28. Fischer, A.H.L., Henrich, T., and Arendt, D. (2010). The normal development of

112

Platynereis dumerilii. Front. Zool. 7, 1–39.

29. Davidson, E.H., and Erwin, D.H. (2006). Gene regulatory networks and the evolution of

animal body plans. Science. 311, 796–800.

30. Erwin, D.H., Laflamme, M., Tweedt, S.M., Sperling, E.A., Pisani, D., and Peterson, K.J.

(2011). The Cambrian conundrum: early divergence and later ecological success in the

early history of animals. Science. 334, 1091–7.

31. Maas, A., Braun, A., Dong, X.P., Donoghue, P.C.J., Müller, K.J., Olempska, E., Repetski,

J.E., Siveter, D.J., Stein, M., and Waloszek, D. (2006). The “Orsten”-More than a

Cambrian Konservat-Lagerstätte yielding exceptional preservation. Palaeoworld 15, 266–

282.

32. Yin, L., Zhu, M., Knoll, A.H., Yuan, X., Zhang, J., and Hu, J. (2007). Doushantuo

embryos preserved inside diapause egg cysts. Nature 446, 661–663.

33. Aria, C., and Caron, J.-B. (2017). Burgess Shale fossils illustrate the origin of the

mandibulate body plan. Nature 545, 89–92.

34. Nanglu, K., Caron, J.-B., Conway Morris, S., and Cameron, C.B. (2016). Cambrian

suspension-feeding tubicolous hemichordates. BMC Biol. 14, 9.

35. Peterson, K.J., Dietrich, M.R., and McPeek, M.A. (2009). MicroRNAs and metazoan

macroevolution: Insights into canalization, complexity, and the Cambrian explosion.

BioEssays 31, 736–747.

113

Figures

114

Figure 1. General morphology of Kootenayscolex barbarensis from the Burgess Shale (Marble Canyon locality). All images oriented with anterior directed to the top of the page. A: holotype (ROMIP64388), nearly complete specimen (posterior missing) showing well preserved palps, median antenna, and chaetae. B-C: paratype (ROMIP64389); B: anterior section showing well-preserved internal head features and sediment infill within the gut; C: elemental map showing phosphatized areas, interpreted as possible remnants of vascularized tissues, within the palps, head and basal portion of parapodia. A thin carbon line running from the front of the head to the palps represents putative neural tissues (see close up Figure 2A-B). D-E: paratype (ROMIP64390); D: paratype alongside putative K. barbarensis juvenile specimens; E: detail showing parapodia preserved in darker black. F-G: paratype (ROMIP62972); F: specimen showing varying parapodial angles and proximal portions of chaetae; G: close up anterior section. H: two paratypes (ROMIP64391.1(left)-2(right)), two complete specimens with elongate chaetae, median antenna and palps. I-J: picture showing seven K. barbarensis specimens, including one paratype (ROMIP63099.1 - see star on figured specimen); J: close up showing a putative juvenile of K. barbarensis preserved with long chaetae. K-N: paratype (ROMIP64392.1-right); K: overall view of paratype with palps, median antenna and mud-filled gut extending through the entire body (paratype partially overlaying a second specimen to the left); L: close up of anterior gut section; M: close up of posterior gut section; N: close up of the left palp showing distal flattening. All images using cross-polarized light except C. Acronyms: an?- anus?, che- chaetae, det- degraded tissue, gin- gut infill, gut- gut, juv- juvenile, mot- mouth, nec- neurochaetae, nep- neuropodia, net- neural tissue, noc- notochaetae, nop- notopodia, mea- median antenna, pal- palp, par- parapodia, pco- prostomial coelom, pep- peristomial parapodia, pec- peristomial chaetae, ppe- proto-peristomium, pro- prostomium,. Scale bars = 1 mm. See also Figures S1, S2, S3.

115

116

Figure 2. Head, parapodial and chaetal morphology of Kootenayscolex barbarensis from the Burgess Shale (Marble Canyon locality). A-C: specimens showing putative location of mouth and internal tissues within palps and median antenna (C); A-B: close ups of boxed area in Fig.1B using direct light (A) and a composite line drawing of 1B and 1C (B); C: paratype (ROMIP64393), superimposed images of both part and counterpart using Apply Image and overlay blending mode in Adobe Photoshop CS6. D-H: specimens showing palps, median antenna (except G) and peristomial chaetae directed anterio-laterally; D: close up of boxed area in Fig. 1A. E: line drawing of 2D. F: paratype (ROMIP64394). G: line drawing of 2F. H: paratype (ROMIP64395). I: close up of boxed area in Fig. 1H. J: close up of chaetal bundle. K: ROMIP64396. L: close up of chaetal bundle. M: ROMIP64397. N: close up of chaetal bundle. Cross-polarized light images (A-H, K, M); scanning electron microscope images (I, J, L, N). For acronyms see Fig. 1. Scale bars = 1 mm. See also Figures S1, S2, S3.

117

118

Figure 3. Anatomy of Kootenayscolex barbarensis from the Burgess Shale. A: oblique view of the head. Dashed lines indicate cut-away transverse cross sections. Red: vascular tissue, blue: putative neural tissue, yellow: mouth and gut. For acronyms see Fig. 1. B: Life reconstruction. Image © Royal Ontario Museum, Danielle Dufault. See also Figures S1, S2, S3.

119

120

Figure 4. Evolutionary implications of Kootenayscolex barbarensis. A: Phylogenetic position of K. barbarensis using majority rules Bayesian analysis (parameters are detailed in Star Methods). Numbers at nods are posterior probabilities. Orange: taxa historically considered part of the major clade “Aciculata”; blue: taxa historically considered part of the ecomorphotype “Sedentaria”; dark blue: Cambrian taxa; pink: sipunculids; green: mollusks. C-B: Two scenarios for annelid head evolution which invoke modern developmental plasticity. B: modern polychaete head arising from a hypothesized ancestor with a biramous peristomium losing the notopodia and notochaetae to produce a uniramous Cambrian condition, then losing the first neuropodia and neurochaeta, leading to the crown group Annelida. C: alternatively, the uniramous peristomium of K. barbarensis and C. spinosa may have been produced during ontogeny (see text for descriptions of each numbered step). See also Figure S4.

121

Appendices

122

Appendix 1. Kootenayscolex barbarensis paratype (ROMIP64389) from the Burgess Shale

Related to Figures 1,2 and 3.

123

124

Appendix 2. Kootenayscolex barbarensis from the Burgess Shale (Marble Canyon locality).

A: full image of paratype (ROMIP64393) preserved in dorsal-ventral orientation and showing parapodial morphology as well as extensive gut infill; the head of a second unprepared specimen is also visible on the same slab. B: image of holotype (ROM 64388) under wet conditions. C-D: full images of paratypes (ROMIP64394 – C; ROMIP64397 – D) showing bundles of chaetae preserved only at the anterior end. E-F: paratype (ROMIP64395); E: full image of specimen before preparation. F: close up of framed area in S2E after preparation showing wide, fanned orientation of prostomial chaetae and base of median antennae merging at the level of base of palps within the protostomium. G: close up of paratype (ROMIP62972) showing variations in orientations of the chaetae as a result of variations in angle of burial, and phosphatized areas at the base of both noto- and neuropodia. All images using cross-polarized light. Superimposed images in D of both part and counterpart using Apply Image and overlay blending mode in

Adobe Photoshop CS6 (see thin dashed line for limits between the two parts). Scale bars = 1 mm. Related to Figures 1,2 and 3.

125

126

Appendix 3. Kootenayscolex barbarensis from the Burgess Shale (Walcott Quarry) and other Cambrian polychaetes from the Burgess Shale showing putative peristomial parapodia and chaetae. A-D: single slab showing two relatively complete specimens

(ROMIP57190.1-2). B: close up of framed specimen in S3A with elongate palps. C: close up of framed specimen in S3A showing median antenna and dark patches at the base of parapodia reminiscent of the Marble Canyon specimens. D: counterpart of S3C. E-F: most complete specimen showing both palps, median antenna and chaetae (ROMIP64399). F: close up of anterior portion of S3E. G-H: Canadia spinosa (lectotype USNM57654), overall view; H: close up of head showing that the first parapodia and chaetae occur posterior to the prostomium. I-K:

Burgessochaeta setigera; I-J: previously unfigured specimen (ROMIP64398) showing that the first parapodia and chaetae also occur posterior to the prostomium, as in C. spinosa (see detail in

J). K: lectotype of B. setigera (USNM57670) which resembles the proposed morphology of

Phragmochaeta canicularis, as a result of both ends of the protostomium being buried into the matrix. Note: posterior end almost indistinguishable from anterior end. Polarized light imagery

(A-F, H-J); direct light imagery (G, K). Scale bars: A=5 mm; B-F: 1 mm; G, H= 2 mm; I, K = 1 mm; J = 0.5 mm. Related to Figures 1,2 and 3.

127

128

Appendix 4. Strict consensus parsimony analysis using our updated character matrix to reflect: (1) the description of Kootenayscolex barbarensis, and (2) re-interpretations of head anatomy of

Canadia spinosa, Burgessochaeta setigera, and Phragmochaeta canicularis (see Star Methods). The four aforementioned Cambrian taxa form a polytomy with all crown group Annelida (excluding sipunculids), Guanshanchaeta felicia, and Pygocirrus butyricampum. This large polytomy is weakly supported. Numbers above nodes are Jackknife support values. Numbers below nodes are Bremer support values. Related to Figure 4.

129

Appendix 5

#NEXUS [written Tue Oct 24 11:09:53 EDT 2017 by Mesquite version 3.11 (build 766) at PAL- Students1/10.20.3.11]

BEGIN TAXA; DIMENSIONS NTAX=80; TAXLABELS Amphiporus Ilyanassa Chaetopleura Wiwaxia Halkieria Crania Lingula Macrochaeta Paralvinella Ampharete Eurythoe Aphrodita Apistobranchus Arenicola Arkonips Burgessochaeta Canadia Notomastus Chaetopterus Dysponetus Cirratulus Cossura Ophryotrocha Esconites Eunice Euphrosine Pherusa Glycera Goniada Guanshanchaeta Leocrates Kenostrychus Lumbricus Lumbrineris Magelona Clymenella Mazopherusa Myzostoma Nephtys Alitta Drilonereis Nothria Ophelia Scoloplos Aricidea Pectinaria Phragmochaeta Eulalia Sigambra Lepidonotus Pygocirrus Kootenayscolex Sabella Serpula Scalibregma Sthenolepis Archaeogolfingia Cambrosipunculus Golfingia Phascolopsis Phascolosoma Sipunculus Sphaerodoropsis Laonice Syllis Amphitrite Trochochaeta Owenia Tubifex Urechis Panthalis Siboglinum Dryptoscolex Fossundecima Sabellaria Protodrilus Saccocirrus Mesonerilla Polygordius Protodriloides ;

END;

BEGIN CHARACTERS; DIMENSIONS NCHAR=190; FORMAT DATATYPE = STANDARD GAP = - MISSING = ?;

MATRIX Amphiporus 00-000-010?00000-0010000?0??0-?????0-00------00----00------0------0------?------0-?000--1000---10-0--00------0000---0-1?000--?001001000100000-00-0--10- 000?00????- Ilyanassa 00-000--1011?100-0010100000?1011100--00------00----00------0------0------0-0000--1000---00-???10------0000---0-00?00--0011001010100000-00-0-- 110011200????- Chaetopleura 000000--10100000-0010100000?1010100--00------00----00------111------00------0-0000--1000---10-???10------0000---0-00?00--00110010101?0000-00-0-- 110011000????- Wiwaxia ??11?--0??????????????0???????????????0----???????????????00----00------10120----10?00--0--0-----0----?0-??00-11000---??????10------???0?????????0000?????0??1?????00- 00-0--?????0-??????? Halkieria ??11?--0??????????????0???????????????0-?-????????????????????????------11120----00?00--0--0-----0----?0-???0-0???0---?????????????????????0?????????0000?????0????????00- ?0-0--?????10??????? Crania 001100-0??00-000-0000000000?0-?????--00------00----00------10020---- 00--11-0--0--0--0---0-0-0000-112-0---10-0--00------0-00---0-00200--000?010-00010000-00-0-- 000101100????- Lingula 001000-0??00-000-0000000000?0-?????--00------00----00------10020--- -00--11-0--0--0--0---0-0-0000-012-0---10-0--00------0-00---0-00200--00??010-00010000-00-0-- 000101100????-

130

Macrochaeta 1111100010????????????00000?10?????21111140???????????????00- 100001100--1000110-100-01100--0-----0--1000-?000-0100100-11?12000------00?011- 01011?01000?01000-00100100-00-0--111000-?0????0 Paralvinella 1111100010100??0-?????00100?10???001100------011-?- 0001010110000----01001100--1010100---000-?000-0110100-11?12000------001011-10- 11000000100-00-00100100-00-0--111000-00????0 Ampharete 1111100010110??0-101??00000?10???001100------011-?- 0001010110000----01001100--1110101100000-1000-0110100-1??12000------000011- 1101100000010??00-00100100-00-0--111000-0001001 Eurythoe 1111100010111100-111??00000?10???00000111010001100000101--00- 000110000--11000----00-01111110-----0---0010101101100100-11012100------0010100- 1210000000011100-00100100-00-0--111000-101?010 Aphrodita 11111100??10???101?1??00??0?10???00000111010001110010100--00- 000110010--10000----00-01111110-----0---0111001101101100-10-111010010------10010--- 1010?00000001100-00101100-00-0--111000-000?110 Apistobranchus 1111100010????????????00000?10???0010011140000001111000-0-00- 100000000--10010----00-01011110-----0---000-11011010010??1??12000------00100--- 1110?00000001000-001??100-00-0--111000-?00?110 Arenicola 111110001010?000-1111000000?10???001100----???????????????00--?- 0011010110000----00-01100--1011100--0000-0000-0100101011111100------0010100- 1010010000100-00-00100100-00-0--111000-00????0 Arkonips ??11?100??????????????0?????????????0011101???????????????00-00?110000-- 1?00??????0-01?????0??????????010???1?01??100-?????????????????????00--- ?????0000?????0??0????100-?0-0--??-000-??????? Burgessochaeta ??11?000??????????????0?????????????00111?1???????????????00- 00?000000--10000----00-01110--0-----0---000-??00-010000?-???1?000------0??00--- ??1??0000?????0??0????100-00-0--???000-??????? Canadia ??111000??????????????0?????????????00111?1???????????????00-00?002000-- 10000----10-01010--0-----0---000-??00-110000?-???12000------0??0101-?????0000???0- 0??0????100-00-0--???000-??????? Notomastus 1111100010111000-1111001001010???00?000----???????????????10--?- 0011011110000----00-01100--1100110--0100-1000-0100101011111000------00000--- 1000010000101000-00100100-00-0--111000-00????0 Chaetopterus 1111101010?10000-0010?00000010???0010011140000001100010-0-00- 1000001010110000----01001101101010100--0000-?000-0100100-1100--00------0?00--- 1111000000001000-00100100-00-0--111000-0?01010 Dysponetus 11111100101000010111??01100?10???00?0011101???????????????00- 000112000--10001010010-01101110-----11010010001101100100-10-111011010------10000--- 101010000000??00-0010?100-00-0--111000-?0????0 Cirratulus 11111000101000011110??00000?10???0021011140???????????????00- 100001100--10000----00-01110--1100100--0000-0000-0100101?11012000------001011- 01011000000001000-00100100-00-0--111000-000?000 Cossura 1111100010????????????00000?????????00??---???????????????00----001100-- 10000----00-01110--0-----0---000-00011010010101??1??00------000011-11010?0000?001?00- 0010?100-00-0--111000-?0????0 Ophryotrocha 11111000??110??111111001000??0110??00011111???????????????00-00011- 010--10011011000-01-111111001-0--10010?01100100101011012101401010000000100--- 1010000?00001100-0010?100-00-0--111000-10????0 Esconites ??11?000??????????????0?????????????0011111???????????????00-00?11-010-- 1?01?????00-01-1111??????0--??010??1100100101????1??014?0103?1?1???0100- ?????00?0?????0??0????100-00-0--???000-???????

131

Eunice 1111100010110??11111??00000?10???000001111100011111000100000-00011- 010--10011011000-01-111111001-0--11010101110100101110-1210140010311110010100- 1010000100001100-00100100-00-0--111000-0010010 Euphrosine 11111000?????????1????00??0?10???0000010---???????????????00--?-110000-- 11000----00-01111110-----0---0010001101100100-1??12100------00?0100-1210000000011100- 00100100-00-0--111000-?0????0 Pherusa 111110001?100??0-10???00000?10???00211111400001110100101--00- 100001100--1000110-100-11110--0-----1100000-?000-0110100-11012000------001011- 11011101000001000-00100100-00-0--111000-?000110 Glycera 11111000??1000010011??0???0?0-0010?00011101000000011011---00- 000010010--10001010000-01101110-----0---0010001100100110-111111013110------1000100- 10000000-1000-00-00100100-00-0--111000-0010110 Goniada 11111000??100??1001???00??0?0-0010?10011101000000011011---00- 000010010--10001010000-01101110-----0---0010001100101110-110111013?10------10000--- 1?0000000000??00-00100100-00-0--111000-?010110 Guanshanchaeta ??11?000??????????????0?????????????00111?????????????????00-??- 000000--1000?????00-01110--0-----0---?00-??0100100?0?????1??00------???0?--- ?????0000?????0??0????100-00-0--???000-??????? Leocrates 111111001010000100111000000??0001??00011101000001111000-0-00- 010110010--10001010000-01101110-----0---0010001100100100-110111011010------10000--- 1010000001001100-00101100-00-0--111000-1000110 Kenostrychus ??11?000??????????????0?????????????0011101???????????????00- 00?110000--1??0?????00-01??1?????????????010??1??010?100-?????100------???0101- ?????0000?????0??0????100-?0-0--???000-??????? Lumbricus 1111100011?00??0-0?01011101110???01?000----???????????????00--?-001100- -10000----00-01110--0-----0---000-0000-0100101011113000------00000---0-10000000001000- 00100100-00-0--?11000-1001??0 Lumbrineris 1111100010110??11101??00000?10???0000010---???????????????00-00011- 010--10011011000-01-111111001-0--1101010010010010111??12101401011110100100--- 10??000000000-00-00100100-00-0--111000-?0????0 Magelona 1111100010100000-001??0000?0?-?????110111200001100000101--00- 200000000--10000----00-01110--1100110--0100-10010010010??11012000------00100---0- 1020000?000000-00110100-00-0--111000-1100110 Clymenella 1111100010100?00-1011000000?10???001000----???????????????00--?- 0011010110000----01001100--1101100--0000-000110100100-11112000------00100--- 1010?00000101000-00100100-00-0--111000-?0????0 Mazopherusa ??11?000??????????????0??????????????1111?????????????????00- ???001100--1000?????00-11??0--??????????000--000-0???10????????00------???011- 0?????0100??0??0??0????100-00-0--???000-??????? Myzostoma 111110000-11?00??111??0101100-1110?000??---???????????????00--?-00- 010--10010----00-01011010-----110000000010-0101100?11011100------10?00---0-00?00000000-00- 00100100-00-0--111000-?0????0 Nephtys 11111100??1000?10101??00??0?0-0010?00011101???????????????00- 000010000--10000----00-01111110-----0---0010001100100100-10-111011010------1000101- 1010000001001100-00101100-00-0--111000-?0????0 Alitta 1111110010100????1011000001?10???0000011101000001111110-0-00- 010010010--10001110000-00001110-----11110010001100100100-110111011010------10000--- 1010000001001100-00101100-00-0--111000-1010110 Drilonereis 11111000??????????????00????????????00????????????????????00--?-11-010-- 10010----00-01011110-----0----01010011010110111??12101401112011000100---1010?0000?001100- 0010?100-00-0--1?1000-?0????0

132

Nothria 111110001011000111011?00000?10???000001111100011111000100-00-00011- 010--10011011001001-111111001-0--1101010110010010111101210140010311110010100- 1010000100001100-0010?100-00-0--111000-?01?110 Ophelia 1111100010111?00-111??00000?10?????00010---0001110010000--10--?-001100- -10000----00-01110--0-----0---000-100110100101010-11000------0010100- 1011010000101001100100100-00-0--111000-00???00 Scoloplos 1111100010110??111101000??201011?0000010---000001001000---00--?- 000000--10000----00-01111111100100--00000110110100101011012000------001011- 0101?000000001100-00100100-00-0--111000-1010000 Aricidea 1111100010100000-110??00000??0?????10010---000001001000---10--?-101100- -10000----00-01110--1100100--01000100110100100-11?1?000------001011-01010?00000001000- 00100100-00-0--111000-?000?00 Pectinaria 1111101010100000-111??00000010???101100------010-?- 0011010110000----11101100--1010111110000-1?00-0110100-11?12000------001011- 11111000000101000-00100100-00-0--111000-0001001 Phragmochaeta ??111?????????????????0??????????????0????-???????????????????????0000- -10000----00-01110--0-----0---000-??0??1100????11????00------??- 0?????????00?0?????0??0?????0??00-0--???000-??????? Eulalia 111111001010000101111?0000000?0010?00011101000001001000---00- 000110010--10001010000-01111010-----0---0010001100100100-11011100------10000--- 1000000000001100-00100100-00-0--111000-1010000 Sigambra 11111100??100001010???00????????????1011101???????????????00- 000110010--10010----00-01111110-----0---0010001100101100-1??11100------10000--- 1000?01000000-00-00100100-00-0--111000-?0????0 Lepidonotus 1111100010100011001110000000100110000011101000101001010---00- 000110010--10000----00-01111110-----11000011001100101100-10-111010110------10010--- 1010?00000001100-00101100-00-0--111000-1000110 Pygocirrus ??11?000??????????????0?????????????00111?????????????????00-???000000-- 10000----00-01110--0-----0---000-??0100100?0????????00------???00---?????0000?????0??0?????00- 00-0--???000-??????? Kootenayscolex ??11?000??????????????0?????????????00111?1???????????????00- 00?100000--10000----00-01110--0-----0---000-??00-1100001-?1????00------???00--- ??1??0000?????0??0????100-00-0--???000-??????? Sabella 1111101010110?00-010??00000?11???102001103111111111101010100- 1000011010110000----01001110--1010110--0000-0000-010010121000--00------0000--- 1010100010001000-00100100-00-0--111000-0000101 Serpula 1111101010111100-0111000000011101102001103111111111101010100- 1000001010110000----01001110--1010110--0000-0000-010010121100--00------0000--- 1010100010001000-00101100-00-0--111000-000011? Scalibregma 11111000??????????????00??0?10???00000111?1000001001000---00- ?0?001100--10000----00-01110--0--?111110000-100110100101011111000------0000100- 1010010000101000-00100100-0--0--111000-0000100 Sthenolepis 11111100101000110??11000000??0???0000011101000101001010---00- 000110010--10001010000-011111111001111110011001100101100-10-111010110------10010--- 1010?00000001100-00101100-00-0--111000-?000110 Archaeogolfingia ??-0?0--??????????????0?????????????-00----???????????????00--?-00------000------0------0-0--1200---???0--00------?000---?????00???????????????11??0110-- ??-000-??????? Cambrosipunculus ??-0?0--??????????????0?????????????-00----???????????????00--?-00------000------0------0-0--1200---???0--00------?000--- ?????00???????????????1010011111??-000-???????

133

Golfingia ?0-000--??110000-001??0000??10?????--00------00--?-00------000------0------0-0--1200---1100--00------0000---1?00200--0001000-00100111011?10011-000- 100???0 Phascolopsis 10-000--10110000-001??00000?10???10--00------00--?-00------000------0------0-0--1200---1100--00------0000---1?00200--0001000-0010010101100--11- 000-100???0 Phascolosoma 10-000--10110000-001??00000?10???10--00------00--?-00------000------0------0-0--1200---1100--00------0000---1?00200--0001000-00100101111?11011- 000-100???0 Sipunculus ?0-000--??110000-001??0000??10?????--00------00--?-00------000------0------0-0--1200---1100--00------00?0---1?00200--0001000-00100101011?0--11-000- 100???0 Sphaerodoropsis 11111100??????????????0???0?000010010011101------00-000111010- -1001101?000-01011010-----1101000-001100100100-10-11100------11000---101??01000001100- 0010?100-00-0--111000-?0????0 Laonice 1111100010100000-1011000000110???001001114000011110111100-00- 100101100--10000----00-01110--1100110--0100-110110100100-10-12000------001011- 01110000000001000-00100100-00-0--111000-1100110 Syllis 1111110010100??111111?000000?01????00011101000001001000---00- 000111000--10001111000-011110111001-11010010001100100100-110111012-10------11000--- 1010000000001100-00100100-00-0--111000-1000110 Amphitrite 1111110010100??0-11110000000101?100?100----???????????????010-?- 0011010110000----01001100--1110100--0000-0000-0110100-11?12000------001011- 11?11000000101000-00100100-00-0--111000-?001001 Trochochaeta 11111000??100000-10???00??0?10????01001114000011110100100000- 100000000--10000----00-01110--0-----1100000-110110101100-1??11000------00000--- 11???00000?01000-0010?100-00-0--111000-1100110 Owenia 1111101010111100-001??0000001000010?0011121???????????????00- 1000011010110000----01001100--1100100--0000-000100100101010-12000------00100---0- 1000000?001000-00110100-00-0--111000-00????1 Tubifex 1111100011?00??0-0?01011101110???01?000------00--?-001100-- 10000----00-01110--1100010--0000-0000-0100101011113000------00000---0-10000000001000- 00100100-00-0--?11000-1001??0 Urechis 110110011011110??11110000000100?000??00------00--?-00-100-- 10000----00-00--0--0-----11-0000-0000-011010??011???00------00000---0-00010000000-00- 00100100-00-0--111000-?00???0 Panthalis 11111100??????????????00????10?????0001110100011100101000-00- 000110000--10000----00-01111110-----0---0111001100101100-10-111010110------10010--- 10???00000001100-00100100-00-0--111000-?00?110 Siboglinum 1111100010100??0-11???00000?1??????20011?40???????????????00- 1000011011110000----01001110--1010110--0000-0000-00--10101100--00------0?000---0- 11000000000-00-00110100-00-0--111000-00???00 Dryptoscolex ??11??00??????????????0?????????????0011101???????????????00- 00?010010--100010???00-0111111????????????11???100???10?????1??010?10------???10--- ?????0000?????0??0????100-00-0--??1000-??????? Fossundecima ??11?100??????????????0?????????????0011101???????????????00- 00?1?0000--1000??????0-0111111????????????10???10010?10?????1??011?10------???00--- ?????0000?????0??0????100-00-0--???000-??????? Sabellaria 1111101010100000-111??0000001001100200111401111101100000--00- 1000010011010000----11101110--1010110--0000-0000-0100100-1??0--00------000011- 11011000010001000-00100100-00-0--111100-0000100

134

Protodrilus 11--100010110000-111??00001?0?100??1001114100011110100001000-10100-- 00--0------0------0-?00100100101011?12000------0100---1010?000- 0001001101110100-00-0--11-000-1101000 Saccocirrus 11111010??11?000-1111?00??0?10???000-01114100011101100001000-101001- 00--10020----00-011-0--0-----0---000-00010010010101??0--00------0100--- 1010?00000001001101110100-00-0--111000-1?0???0 Mesonerilla 1111100010100??0-111??00000?00??????0011101000001001000---00- 000110000--1000100-000-01110--0-----0---0010000100100100-1??12000------0?100--- 1010?0000???110100010?100-00-0--111000-100?1?0 Polygordius 11--10-010111100-101??0?00?010111000-011121000001000000---00-00000-- 00--0------0-00010-100101011????------0-100---1010?000-0001001100010100- 00-0--11-000-10????0 Protodriloides 110110?00-10???0-000??0002??00???0?0001112100011100100000000- 10100??00--100?0----00-00000000------000-000100100101010?12000------00100--- 1010?00000000-011011?0100-00-0--11-000-1?????0

;

END; BEGIN ASSUMPTIONS; TYPESET * UNTITLED = unord: 1 - 190;

END;

BEGIN MESQUITECHARMODELS; ProbModelSet * UNTITLED = 'Mk1 (est.)': 1 - 190; END;

Begin MESQUITE; MESQUITESCRIPTVERSION 2; TITLE AUTO; tell ProjectCoordinator; timeSaved 1508857793399; getEmployee #mesquite.minimal.ManageTaxa.ManageTaxa; tell It; setID 0 4465894558552123987; endTell; getEmployee #mesquite.charMatrices.ManageCharacters.ManageCharacters; tell It; setID 0 9179421517133166942; tell It; setDefaultOrder 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 189 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188; attachments ;

135

endTell; mqVersion 311; checksumv 0 3 1335762284 null getNumChars 190 numChars 190 getNumTaxa 80 numTaxa 80 short true bits 31 states 31 sumSquaresStatesOnly 24411.0 sumSquares 24411.0 longCompressibleToShort false usingShortMatrix true NumFiles 1 NumMatrices 1; mqVersion; endTell; getWindow; tell It; suppress; setResourcesState false false 100; setPopoutState 300; setExplanationSize 0; setAnnotationSize 0; setFontIncAnnot 0; setFontIncExp 0; setSize 1620 900; setLocation 109 15; setFont SanSerif; setFontSize 10; getToolPalette; tell It; endTell; desuppress; endTell; getEmployee #mesquite.charMatrices.BasicDataWindowCoord.BasicDataWindowCoord; tell It; showDataWindow #9179421517133166942 #mesquite.charMatrices.BasicDataWindowMaker.BasicDataWindowMaker; tell It; getWindow; tell It; setExplanationSize 30; setAnnotationSize 20; setFontIncAnnot 0; setFontIncExp 0; setSize 1520 828; setLocation 109 15; setFont SanSerif; setFontSize 10; getToolPalette; tell It; setTool mesquite.charMatrices.BasicDataWindowMaker.BasicDataWindow.ibeam; endTell; setActive; setTool mesquite.charMatrices.BasicDataWindowMaker.BasicDataWindow.ibeam; colorCells #mesquite.charMatrices.NoColor.NoColor;

136

colorRowNames #mesquite.charMatrices.TaxonGroupColor.TaxonGroupColor; colorColumnNames #mesquite.charMatrices.CharGroupColor.CharGroupColor; colorText #mesquite.charMatrices.NoColor.NoColor; setBackground White; toggleShowNames on; toggleShowTaxonNames on; toggleTight off; toggleThinRows off; toggleShowChanges on; toggleSeparateLines off; toggleShowStates on; toggleAutoWCharNames on; toggleAutoTaxonNames off; toggleShowDefaultCharNames off; toggleConstrainCW on; toggleBirdsEye off; toggleShowPaleGrid off; toggleShowPaleCellColors off; toggleShowPaleExcluded off; togglePaleInapplicable on; toggleShowBoldCellText off; toggleAllowAutosize on; toggleColorsPanel off; toggleDiagonal on; setDiagonalHeight 80; toggleLinkedScrolling on; toggleScrollLinkedTables off; endTell; showWindow; getWindow; tell It; forceAutosize; endTell; getEmployee #mesquite.charMatrices.AlterData.AlterData; tell It; toggleBySubmenus off; endTell; getEmployee #mesquite.charMatrices.ColorByState.ColorByState; tell It; setStateLimit 9; toggleUniformMaximum on; endTell; getEmployee #mesquite.charMatrices.ColorCells.ColorCells; tell It; setColor Red; removeColor off; endTell; getEmployee #mesquite.categ.StateNamesStrip.StateNamesStrip; tell It;

137

showStrip off; endTell; getEmployee #mesquite.charMatrices.AnnotPanel.AnnotPanel; tell It; togglePanel off; endTell; getEmployee #mesquite.charMatrices.CharReferenceStrip.CharReferenceStrip; tell It; showStrip off; endTell; getEmployee #mesquite.charMatrices.QuickKeySelector.QuickKeySelector; tell It; autotabOff; endTell; getEmployee #mesquite.charMatrices.SelSummaryStrip.SelSummaryStrip; tell It; showStrip off; endTell; getEmployee #mesquite.categ.SmallStateNamesEditor.SmallStateNamesEditor; tell It; panelOpen true; endTell; endTell; endTell; endTell; end;

BEGIN MRBAYES; Set autoclose=yes; Set nowarnings=yes; Log start filename= NC_reversed1.log replace;

Outgroup Amphiporus;

Prset statefreqpr=dirichlet(500); Lset coding=variable rates=gamma; Showmodel;

Mcmc ngen=10000000 samplefreq=1000 printfreq=100 nchains=4 temp=0.2 nruns=2 Filename= NC_reversed1.out burninfrac=0.25 starttree=random savebrlens=yes; Sump filename=NC_reversed1.out burninfrac=0.25 nruns=2; Sumt filename=NC_reversed1.out burninfrac=0.25 Contype=Halfcompat;

138

END;

139

New Burgess Shale polychaete reveals the origin of the annelid head - Figured Specimens of Kootenayscolex barbarensis- Nanglu and Caron 2018

ROM No. Level (-cm) Locality Type Field No. # of Specimens Figures Supplementary information 64388 372 Marble Canyon Holotype 2016-0833 2 1a,2e S2b 64389 378 Marble Canyon Paratype 2012-0221 3 1bc,2ab S1abcdefg 64390 405 Marble Canyon Paratype 2014-1010 52 1de 62972 381 Marble Canyon Paratype 2012-0594 4 1fg S2g 64391.1 378 Marble Canyon Paratype 2012-0206 1 1h,2ij 64391.2 378 Marble Canyon Paratype 2012-0206 2 1ij,2df 63099.1 393 Marble Canyon Paratype 2012-0719 9 1klmn 64392 407 Marble Canyon Paratype 2014-0971 9 2c S2a 64393 378 Marble Canyon Paratype 2012-0185 4 2g S2c 64394 377 Marble Canyon Paratype 2016-0811 1 2h S2ef 64395 393 Marble Canyon Paratype 2012-0721 7 2kl 64396 381 Marble Canyon Paratype 2012-0576 1 2mn S2d 64397 383 Marble Canyon Paratype 2014-0769 1 4cd 57190 350 Walcott Quarry Paratype 1999-1690 2 S3abcd 64399 350 Walcott Quarry Paratype 2000-2113 1 S3ef Total 99

140

Other ROM specimens (unfigured)

Marble Marble Canyon stratigraphic abundances Walcott Quarry Canyon Field No. Level (-cm from Eldon contact) Number of specimens Intervals (cm below Eldon Fm. contact) Abundance Field No. Level (cm) 2012-0153 327 1 251-260 2 2000-1277 -350

2012-0216 378 2 261-270 2 2000-1671 -350

2012-0217 378 1 271-280 0 2000-1740 -350

2012-0219 378 2 281-290 0 2000-2069 -350

2012-0220 378 1 291-300 3 1999-1245 -350

2012-0222 378 3 301-310 1 1999-1652 -350

2012-0225 378 1 311-320 0

2012-0226 378 2 321-330 4

2012-0227 378 2 331-340 1

2012-0228 378 1 341-350 11

2012-0229 378 3 351-360 8

2012-0230 378 2 361-370 7

2012-0231 378 1 371-380 79

2012-0232 378 1 381-390 53

2012-0234 378 1 391-400 132

2012-0235 378 1 401-410 88

2012-0237 378 1 411-420 67

2012-0238 378 9 421-430 12

2012-0239 378 1 431-440 0

2012-0241 378 1 441-450 0

2012-0242 378 1 451-460 0

2012-0243 378 4 461-470 0

2012-0245 378 1 471-480 2

2012-0300 378 1 Level unknown 28

2012-0447 talus 1 Total (includes both figured and unfigured specimens) 564

2012-0454 262 1

141

2012-0463 297 1

2012-0465 297 2

2012-0500 348 2

2012-0501 348 1

2012-0502 348 1

2012-0511 347 1

2012-0524 377 1

2012-0525 381 1

2012-0533 356 1

2012-0536 381 1

2012-0548 383 1

2012-0549 381 1

2012-0564 372 1

2012-0567 384 1

2012-0569 383 2

2012-0575 387 2

2012-0583 383 1

2012-0594 381 4

2012-0598 406 1

2012-0599 406 1

2012-0600 406 1

2012-0601 406 1

2012-0602 406 2

2012-0604 406 5

2012-0605 406 2

2012-0606 396 1

2012-0607 406 3

2012-0612 406 2

2012-0614 talus 4

2012-0615 talus 1

142

2012-0619 393 2

2012-0623 407 2

2012-0626 384 1

2012-0627 393 2

2012-0629 393 2

2012-0632 396 2

2012-0633 396 1

2012-0634 396 3

2012-0635 396 3

2012-0636 396 3

2012-0637 396 2

2012-0638 405 1

2012-0641 411 5

2012-0647 412 1

2012-0648 413 1

2012-0650 413 1

2012-0651 406 1

2012-0652 412 1

2012-0653 412 1

2012-0654 410 1

2012-0655 410 1

2012-0657 415 1

2012-0660 402 2

2012-0661 402 1

2012-0667 404 2

2012-0668 403 1

2012-0670 402 3

2012-0672 415 2

2012-0674 399 3

2012-0676 416 2

143

2012-0677 394 1

2012-0678 394 6

2012-0679 394 1

2012-0680 410 2

2012-0681 394 3

2012-0682 393 8

2012-0684 393 12

2012-0688 423 1

2012-0689 410 1

2012-0695 396 1

2012-0696 395 3

2012-0697 395 2

2012-0699 411 5

2012-0704 382 2

2012-0705 416 4

2012-0709 401 2

2012-0714 407 1

2012-0715 422 1

2012-0720 377 4

2012-0723 416 1

2012-0728 353 1

2014-0591 261 1

2014-0594 253 1

2014-0598 260 1

2014-0650 352 1

2014-0659 338 1

2014-0681 385 1

2014-0682 385 1

2014-0700 330 1

2014-0726 383 1

144

2014-0735 383 1

2014-0743 385 2

2014-0746 391 1

2014-0750 385 1

2014-0760 386 1

2014-0795 383 1

2014-0809 387 1

2014-0810 407 1

2014-0821 385 1

2014-0836 413 5

2014-0840 NA 1

2014-0842 403 2

2014-0844 382 1

2014-0845 388 1

2014-0859 347 1

2014-0881 350 1

2014-0896 353 1

2014-0908 354 1

2014-0910 351 2

2014-0911 305 1

2014-0927 363 2

2014-0945 408 2

2014-0954 348 1

2014-0964 348 1

2014-0969 397 2

2014-0970 385 1

2014-0976 407 2

2014-0978 396 1

2014-1001 405 1

2014-1004 405 1

145

2014-1005 412 1

2014-1024 363 2

2014-1025 415 1

2014-1034 414 1

2014-1041 345 1

2014-1043 363 3

2014-1054 416 1

2014-1059 423 8

2014-1064 395 3

2014-1068 410 1

2014-1075 385 3

2014-1077 387 1

2014-1080 385 2

2014-1097 415 3

2014-1102 406 1

2014-1103 415 1

2014-1109 391 2

2014-1110 380 1

2014-1114 386 1

2014-1117 384 1

2014-1123 415 2

2014-1128 417 1

2014-1139 387 1

2014-1140 398 1

2014-1147 398 1

2014-1152 400 1

2014-1155 400 1

2014-1158 386 1

2014-1161 387 1

2014-1191 418 3

146

2014-1199 386 1

2014-1201 380 1

2014-1202 talus 1

2014-1209 418 1

2014-1212 419 1

2014-1214 428 1

2014-1215 417 6

2014-1216 418 1

2014-1223 415 2

2014-1250 374 4

2014-1264 416 1

2014-1272 403 1

2014-1380 353 1

2014-1421 350 1

2016-0714 talus 1

2016-0719 talus 5

2016-0720 talus 5

2016-0725 384 1

2016-0726 384 1

2016-0730 387 1

2016-0732 394 1

2016-0742 404 2

2016-0743 NA 1

2016-0750 NA 1

2016-0753 404 1

2016-0758 393 28

2016-0761 403 10

2016-0765 408 1

2016-0766 417 1

2016-0772 329 2

147

2016-0778 380 1

2016-0779 383 1

2016-0780 375 2

2016-0787 380 1

2016-0790 380 2

2016-0792 384 1

2016-0801 480 1

2016-0806 376 1

2016-0809 375 2

2016-0811 377 1

2016-0812 400 2

2016-0813 400 10

2016-0817 415 1

2016-0819 418 3

2016-0820 381 1

2016-0823 417 1

2016-0830 380 1

2016-0834 380 1

2016-0835 403 4

2016-0836 NA 1

2016-0837 417 1

2016-0840 418 1

2016-0843 NA 1

2016-0848 417 3

2016-0904 425 1

2016-0950 480 1

2016-1010 401 5

2016-1023 380 1

2016-1024 380 1

2016-1025 380 2

148

2016-1026 380 1

2016-1030 381 1

2016-1032 380 1

2016-1036 382 1

2016-1039 378 2

2016-1040 409 1

2016-1042 392 1

2016-1055 NA 5

Total 465

149

CHAPTER 4

Diversity and structure of the Marble Canyon paleocommunity,

Kootenay National Park, Burgess Shale, British Columbia

Karma Nanglu1 and Jean-Bernard Caron1,2,3

1Department of Ecology and Evolutionary Biology, University of Toronto, Toronto, Ontario

M5S 2J7, Canada

2Department of Natural History Palaeobiology, Royal Ontario Museum, Toronto, Ontario

M5S 2C6, Canada.

3Department of Earth Sciences, University of Toronto, Toronto, Ontario, M5S 3B1, Canada

ABSTRACT

Konservat Lagerstätten, sites of exceptional fossil preservation, are key to providing insights on the diversity and structure of ancient ecosystems. The 508-million-year-old Burgess Shale is among the most important such sites, as it provides a direct window into some of the earliest complex animal communities. Here we describe the most recently discovered major Burgess

Shale site, the Marble Canyon fossil locality in Kootenay National Park (British Columbia). A total of 73 species were identified across 16,438 specimens from a within a 4.2m thick stratigraphic interval divided into 10 cm bin intervals. Multivariate methods (correspondence analysis) reveal a temporal division in the main ecological categories over the studied interval,

150 suggesting major ecological and/or environmental changes: an assemblage dominated by suspension feeders at the uppermost levels of the quarry, one dominated by hunter/scavengers in the intermediate strata, and one dominated by deposit feeders at the lowermost levels of the quarry. Many taxa, such as hemichordates, chordates and annelids, are more abundant at Marble

Canyon than at any other Cambrian locality. Observations of faunal turnover throughout the sampled interval were integrated with species abundance data from three other Burgess Shale sites to describe broader geographic and temporal patterns of species composition and ecological trends across the Stephen Formation. The combined Burgess Shale dataset (Marble Canyon,

Walcott Quarry, Raymond Quarry, and Tulip Beds) contains species abundance patterns for

77,179 specimens of 234 taxa. Once analysed quantitatively, these four sites occupy distinct areas in multivariate space with limited overlap suggesting significant differences in their composition. Ecological analyses also suggest that Burgess Shale bedding assemblages from the four localities included can be described as belonging to one of three groups based on the dominant trophic mode, i.e. suspension feeding, macrofaunal heterotrophy, or deposit feeding.

These ecological groupings do not share a direct relationship with the inferred stratigraphic positioning of the four Burgess Shale localities, suggesting that the community was ecologically dynamic with similar functional groups occupied by various sets of species. Taken together, these patterns of taxonomic and ecological change suggest a high degree of temporal and spatial variation in both species assemblages and niche occupation across the Burgess Shale.

Environmental disturbance, as has been hypothesized before, may have played a key role in structuring the community at the bedding assemblage level, while temporally or spatially disjunct localities may represent more long-term changes in biotic or abiotic conditions.

151

1. INTRODUCTION

The 508-million-year-old Burgess Shale (middle Cambrian), located in the Canadian

Rockies in British Columbia, is among the most celebrated Konservat Lagerstätten in the world thanks to the exceptional preservation of abundant and diverse assemblages of soft-bodied animals. For over 100 years, the Burgess Shale fauna has provided unique insights into the morphology and evolutionary relationships of some of the oldest members of many metazoan groups within the broader context of the Cambrian radiation of animal life (Erwin et al. 2011). In addition to its contribution to evolutionary studies, the Burgess Shale offers a far more complete view of Cambrian marine life than do normal fossil deposits of the same age. While other deposits tend to consist of hard-shelled taxa only, in the Phyllopod Bed fauna of the Walcott

Quarry—the first Burgess Shale locality to be discovered by Charles D. Walcott in 1909

(Conway Morris 1986)—the shelly taxa typical of most Cambrian fossil deposits were estimated to be only about 2% of the total community by abundance (Conway Morris 1986), demonstrating starkly the richness of data that such a soft-bodied assemblage could provide. Subsequent paleocommunity studies using a range of quantitative tools have not only shown how diverse and complex the Burgess Shale is, in both composition and life habits of the taxa within it, but also how it varies within the Walcott Quarry itself between particular bedding assemblages (Caron and Jackson 2008). Comparisons with the Tulip Beds on the nearby Mount Stephen, revealed that only half of the genera of the Tulip Beds are shared with the Walcott Quarry (O’Brien and

Caron 2015). Despite these differences, the major taxonomic groups (Arthropoda, Porifera,

Priapulida, Brachiopoda) and ecological modes (epibenthic suspension feeders and nektobenthic predators) were broadly similar at the two localities (O’Brien and Caron 2015).

152

These recent studies have begun to elucidate a broader picture of Cambrian community ecology. However local and regional variations of the Burgess Shale paleoenvironment and community composition at different temporal scales, as well as the factors responsible for those variations, are still poorly known. The lack of a unified dataset incorporating fine scale temporal variations in community composition between multiple Burgess Shale localities has precluded a quantitative assessment of the degree of faunal turnover between Cambrian marine communities.

Consequently, the utility of the exceptional Cambrian fossil record to elucidate the deep-time origins of modern ecological dynamics has been limited.

Here we describe the paleocommunity diversity and structure of the Burgess Shale at the

Marble Canyon fossil locality. This site has already revealed key insights into a number of major animal groups such as the Arthropoda (Aria and Caron 2015, 2017; Aria et al. 2015) Chordata

(Morris and Caron 2014), Lophophorata (Moysiuk et al. 2017), Hemichordata (Nanglu et al.

2016) and Annelida (Nanglu and Caron 2018), and is the second-most intensively sampled

Burgess Shale locality after the Walcott Quarry. Stratigraphic data at the 10 cm scale interval allows for uncommonly high-resolution consideration of temporal change within the community, comparable only to the Walcott Quarry. We assemble the largest species abundance data matrix for the Burgess Shale to date, to compare quantitatively the patterns of community compositional turnover and ecological structural variations between the Marble Canyon, Walcott Quarry (Caron and Jackson 2008; O’Brien and Caron 2015), the Tulip Beds (O’Brien and Caron 2015) and the

Raymond Quarry, the four most extensively sampled Burgess Shale fossil assemblages.

Biostratigraphic evidence (trilobite assemblage and proximity to the Eldon Formation) suggests the Marble Canyon was probably the youngest assemblage of the Burgess Shale community despite sharing numerous species with the stratigraphically older Walcott Quarry (Caron et al.

153

2014). This instead suggests a close temporal association and non-uniform sedimentation rates between the two localities which we test here quantitatively using the full fossil assemblages from both sites. These comparisons provide an exceptional record of ecological to sub- evolutionary variations of the Burgess Shale community in time and space and offer new insights into the complexity of Cambrian marine life more broadly.

2. MATERIALS AND METHODS

2.1 The Marble Canyon locality, data collection and taphonomic considerations

The Marble Canyon fossil site (MC) is located in Kootenay National Park, British

Columbia, Canada, 40 km southwest of the Walcott Quarry in Yoho National Park and was inferred to be the youngest Burgess Shale community based on biostratigraphic evidence (Caron et al. 2014). Field collections were conducted in 2012, 2014, and 2016 totalling 21,687 individual specimens (field observations and collected specimens) representing 74 species. The contact point between the Stephen Formation and the overlying Eldon Formation was used as a reference point for stratigraphic measurements, with specimens measured in negative centimetres down from this contact (Figure 1A). Stratigraphic level was recorded for all specimens included in subsequent quantitative analyses. Fossils were organized into stratigraphic sub-units henceforth referred to as bedding assemblages (BA). Each of these sub-units is 10 cm thick, reflecting our confidence in the accuracy of stratigraphic measurements due to lateral variation in bedding horizons throughout the quarry (± 5 cm). Most fossils are preserved in millimetric-thick shale units, each of which represent a single burial event. Thus the sum of all fossils collected

154 within a BA, typically representing dozens of burial events constitute an induced time average assemblage of unknown duration.

The community data matrix of raw species abundances by BA was pruned of all sub-units with a total specimen count under 299 to prevent possible biases in subsequent analyses that are sensitive to low sample size. 299 was chosen as a minimum threshold for inclusion to allow for comparisons with the Walcott Quarry, where a minimum threshold of 300 specimens/BA was used, as well as for initial rarefaction curves which suggested that 100 specimens was too low of a threshold (100 specimens was used as a minimum specimen threshold in the study of the

Chengjiang localities (Zhao et al. 2013). Clustering analyses using thresholds at 100 specimens/BA, 299 specimens/BA, and 500 specimens/BA showed that broad-scale patterns of faunal turnover at Marble Canyon are not sensitive to varying the threshold of BA inclusion

(Appendix 1). The resultant species abundance matrix included 18 BAs from Marble Canyon, the first of which begins 230 cm below the Eldon-Stephen contact point (Figure 1B). In total

16,438 specimens were included in all analyses of faunal composition, diversity, and turnover at

MC.

Specimens were identified to species level whenever possible and tabulated using established methods (Caron and Jackson 2008; O’Brien and Caron 2015), including undescribed species. Some specimens were identified based on disarticulated or dissociated material, including carapaces of bivalved arthropods such as Tokummia katalepsis, dinocariids such as

Hurdia sp., or myomeres of Metaspriggina sp. In rare instances, the number of specimens from surfaces composed exclusively of shelly taxa such as Haplophrentis carinatus or Ptychagnostus sp., with individuals typically overlapping each other, were estimated.

155

The degree to which the original Marble Canyon community has been affected by transport or decay and how these factors may have varied across different BAs or localities is difficult to assess, although in all cases, specimens would have been entombed through similar types of rapid mudflow deposits of varying intensities (Gaines and Droser 2010), and preserved under similar environmental conditions (Gaines et al. 2008, 2012). Previous quantitative work from the Walcott Quarry has suggested a limited impact of taphonomic biases on the structure of the community (Caron and Jackson 2006). The preservation quality at Marble Canyon of a range of soft-bodied animals representing different body plans is comparable to or better in some instances than what is known from Walcott Quarry (Caron and Jackson 2006, 2008; Caron et al.

2014) therefore suggesting a limited effect of taphonomic factors on overall composition. This conclusion is corroborated by the presence of abundant and pristinely preserved vermiform enteropneusts (Oesia disjuncta) across most of the studied interval. Hemichordates are almost entirely soft-bodied and have been shown to decay extremely quickly (Nanglu et al. 2015; Beli et al. 2017); this timeline suggests that specimens from most BAs in MC underwent little pre- or post-mortem transport or decay.

2.2 Other localities

The Walcott Quarry (WQ) is located on Fossil Ridge between Mount Wapta and Mount

Field and is part of the Walcott Quarry Member of the Stephen Formation (Caron and Jackson

2006). We used the dataset originally published by Caron and Jackson (2008) and updated by

O’Brien and Caron (2015) for our quantitative analyses. This dataset comprised 43,296 specimens, divided stratigraphically into 26 BAs with more than 300 collected specimens each.

156

The Raymond Quarry (RQ) is located 22 metres directly above the Walcott Quarry

(Devereux 2001). I collected 9,539 observations from a 2.3 metre-thick stratigraphic section using 10 cm bins to create BAs roughly comparable with those of the MC and WQ. One hundred and fourteen taxa were identified from this relatively restricted stratigraphic section of the RQ, with dominant taxa being the arthropod Leanchoilia, the sponge Choia, and the priapulid Ottoia.

The Tulip Beds locality on Mount Stephen was described in O’Brien and Caron (2015).

This locality shares roughly 50% of its genera with the Walcott Quarry and is largely dominated by an indeterminate egg-shaped taxon, the tulip animal Siphusauctum gregarium, the alga

Marpolia spissa, and the predator Anomalocaris canadensis. Most critically, the Tulip Beds is part of the Campsite Cliff Shale Member which is significantly older than the Walcott Quarry

(and therefore, by definition, the Raymond Quarry) (Fletcher and Collins 2003; O’Brien et al.

2014). This provides a reference point older than both the Walcott Quarry and Raymond Quarry to compare the Marble Canyon against, in order to corroborate its position as the youngest

Burgess Shale locality based on species abundance patterns (Caron et al. 2014). A total of 7,906 specimens from the Tulip Beds were included in our analyses.

With all four localities combined, our ecological data matrix includes abundance patterns for 77,179 specimens of 234 taxa (Appendix 2).

2.3 Diversity patterns and multivariate analyses

All quantitative data was prepared in Microsoft Excel and analyzed in R using the following packages: vegan, MASS, analogue, dendextend, dplyr, indicspecies, iNEXT, and ggplot2.

157

Rarefaction methods were used to determine the extent to which the MC locality had been adequately sampled to recover the majority of the true total species richness. Initial tests indicated that variable sampling efforts may have influenced our analysis of BA diversity. To account for variable sampling effort, two further methods were used. First, species richness was estimated for all bedding assemblages by interpolating richness at 299 specimens and extrapolating richness at 2,566 specimens using the methods established in (Colwell et al. 2012), then these estimates were plotted against the observed richness for each respective bedding assemblage. Two hundred and ninety nine and 2,566 represent the least- and most intensively sampled assemblages, respectively, and were thus chosen to explore how our view of BA-level diversity may be affected by variation in sampling effort. Second, rarefaction curves were plotted using a uniform sample size of 1,446 (equivalent to the third most sampled BA). This level of sampling effort allowed us to observe the ordering of species richnesses/BA from the richest and poorest BAs by extrapolating up from our under-sampled beds and down from our two most heavily-sampled beds. Whittaker plots (also called rank abundance curves) were also plotted for the three richest and poorest BAs to observe how community structure may be influenced by patterns of evenness/dominance of species abundances.

Cluster analysis was used to identify major groups of sub-units sharing similar species or ecological compositions. Two dissimilarity metrics were used to describe patterns of faunal turnover: the Morisita-Horn overlap index and the Jaccard dissimilarity index. The Morisita-

Horn overlap index is a common method for producing ecological distance matrices and has been demonstrated to be robust even with variation in sample sizes (Magurran 2004; Barwell et al. 2015). It has, however, also been demonstrated to be extremely insensitive to turnover in rare species. For this reason, we also used the Jaccard index which, being a presence/absence

158 turnover metric, is more sensitive to these changes (Koleff et al. 2003; Barwell et al. 2015). This index is also widely used, enabling these data to be compared with other community analyses.

Using multiple metrics of beta diversity and contrasting their results is recommended to capture the full breadth of faunal turnover patterns (Koleff et al. 2003). For ecological clustering, only the Morista-Horn index was used, as there are too few ecological modes represented at Marble

Canyon to accurately discriminate clusters.

Correspondence analysis (CA), a method of ordination, was chosen to visualize diversity patterns among the Marble Canyon sub-units because it maintains fidelity of site-to-site, site-to species, and species-to-species relationships, that is to say, it does not make presuppositions regarding the independent versus dependent variables in the multivariate dataset being analyzed.

Using CA, when correspondence axes are plotted, 1) sites similar in species composition should appear close to each other; 2) species should appear close to sites in which they are abundant, and, 3) species should appear close to other species with similar distributions across sites.

Detrended correspondence analysis was used for the multivariate analysis of the Burgess Shale by species composition due to the detection of possible arch effects in the data during initial tests using correspondence analysis.

Finally, species indicator analysis was used to identify species that may be uniquely characteristic of a given locality. By providing a priori knowledge of how particular sites are grouped, this method calculates an indicator statistic for each species in the community data matrix which takes into account : i) the specificity, that is, likelihood of the sampled site (in this case, BA) belonging to a target site group (in this case, a locality of WQ, TB, RQ or MC) given that the species has been found, and ii) the fidelity, that is, probability of finding the species at any sites belonging to the target site group (De Cáceres et al. 2012).

159

2.4 Ecological groups and functional diversity

The ecological mode was described using a modified version of the three-axis system described in (Bambach et al. 2007). This system posits that ecological mode of the vast majority of marine taxa can be described using three variables: i) the vertical position within the water column they occupied; ii) their motility; iii) their trophic strategy. This framework has been used in other analyses of Cambrian fauna (Conway Morris 1986; Caron and Jackson 2008; Zhao et al.

2013; O’Brien and Caron 2015), and is particularly useful when considering animal communities that incorporate a wide variety of phyla. While admittedly over-generalized in some cases, this approach can accommodate the wide variety of ecologies employed by the considerable disparity of body plans represented by a well-preserved marine locality. We then constructed a ecological- group matrix whereby each taxon at MC was assigned an ecological mode using the three-axis system. This dataset was then analyzed using cluster analyses, CA and DCA (as described in 2.4;

Appendix 3).

RESULTS

3.1. Taxonomic community diversity patterns at Marble Canyon

In total, 73 species are present within the BAs analyzed from MC. Abundant species tend to recur across the studied interval (e.g. Liangshanella, Haplophrentis, Oesia, and Peronopsis) while others have patchier temporal distributions (e.g. Yawunik and Tokummia; Figure 1A).

Kootenayscolex and Primicaris seem particularly abundant in thicker beds (between BA 350 and

160

BA 420; Figure 1A). As is the case in all major studied Burgess Shale sites (Conway Morris

1986; Caron and Jackson 2008; O’Brien and Caron 2012), arthropods are the most abundant and diverse major taxonomic group (Figure 1B). However, in no other such paleocommunity are hemichordates, annelids, or chordates such significant components of the community, at 16%,

5% and 1.3% by abundance, respectively. When considering diversity irrespective of abundance, there is a substantial number of dinocariidids, which constitute roughly 11% of the total species richness, many of which are currently undescribed forms (Figure 1B).

The slope of the rarefaction curve for the Marble Canyon as a bulk assemblage, compared to the Walcott Quarry, Raymond Quarry, and Tulip Beds bulk assemblages, suggests that, while all localities have been sufficiently sampled to capture most of the species richness at a BA level, Marble Canyon has the greatest potential for new species discovery with continued sampling (Figure 2A). To account for differential sampling effort, rarefaction curves for individual bedding assemblages were extrapolated to a uniform sampling effort of 1,466 specimens (the number of specimens collected from the third most intensively sampled bedding assemblage; Figure 2B). While the rank order of species richnesses does change between the observed and extrapolated curves, extrapolating the data to 1,466 specimens produces rarefaction curves that generally fall within the 95% confidence interval of most BAs (Figure 2B). This limits the extent to which the rank-order of richnesses per BA in our extrapolated dataset can be used to compare with the observed rank order of richnesses per BA in the actual observed dataset, as these relationships fall within the boundaries of uncertainty with even moderate extrapolation (Figure 2B).

BAs 350, 380, 390, 400 and 410 had the highest observed species richness, while BAs

300, 310, and 320 had the lowest. Estimating species richness by interpolation of rarefaction

161 curves to a uniform sampling size of 299 specimens suggests that discriminating between BA- level differences in diversity is somewhat sensitive to sampling effort, as the interpolated richnesses at this level of sampling do not reveal any discernible pattern (Figure 3).

Extrapolation to a uniform size of 2,566 specimens suggests a relatively similar pattern to the actual observed pattern of species richnesses. However, with this degree of extrapolation, the species richness of most bedding assemblages falls within the confidence intervals of their adjacent assemblages (Figure 3).

Rank abundance curves illustrate two main types of assemblages. Most BAs have relatively steep curves, indicating that their respective temporal sub-communities were numerically dominated by relatively few taxa (Figure 4). In most cases, these were the four most abundant and stratigraphically wide-ranging taxa, namely Haplophrentis and Oesia, which are stratigraphically widespread but most dominant between BA 230 and BA 330, and

Liangshanella and Peronopsis, particularly below BA 330. BAs with less steep rank abundance curves are indicative of communities that are less dominated by a small group of taxa. In addition to having the greatest evenness of species composition, BAs 380, 400 and 350 are also the most species-rich sub-units (Figures 3, 4).

All BAs at Marble Canyon are dominated by one of three groups: Arthropoda,

Hemichordata, or Lophophorata (Figure 5). Hemichordate and lophophorate dominances are almost entirely due to Oesia disjuncta and Haplophrentis, respectively. The abundance of other species of hemichordate are negligible: Spartobranchus, new hemichordate A and new hemichordate B together make up 0.2% of the community, and the other species of lophophorates (paterinids, acrotretids, lingulids, Micromitra, and Linnarsonnia) make up 1.3%.

Together these three dominant taxa (Arthropoda, Hemichordata, and Lophophorata) constitute

162 between 81% (BA 390) and 99% (BA 230 and BA 510) of the total species abundance. There is an easily observable trend in major taxonomic abundances stratigraphically in MC (Figure 5); the upper levels of the quarry—the youngest strata—are dominated by hemichordates and lophophorates. This trend persists until BA 330, the youngest strata at which arthropods are the most abundant major taxonomic group. This change is due to both the increase in abundance of large predatory arthropods (Yawunik, Sidneyia, Tokummia) and, more significantly, the increase in abundance of the arthropod Liangshanella, the most abundant taxon in BA 350 to BA 420.

Arthropods remain the most numerically dominant taxon for all BA below BA 330.

Taxonomic cluster analyses of the Marble Canyon using both the Morisita-Horn index and Jaccard index distinguish two principal clusters (Figure 6). The Lower Quarry (LQ) consists of BAs 480 to 500. The Upper Quarry (UQ) consists of all other BAs. The UQ is further subdivided stratigraphically in the Morisita-Horn cluster analysis into the uppermost levels of the

Upper Quarry (BAs 240 to 340; UQ1) and the lower levels of the Upper Quarry (BAs 350 to

420; UQ2). The Jaccard cluster analyses broadly agrees with these groupings, except that BA

360 is not grouped with the other UQ2 sites (Figure 6).

The correspondence analysis corroborates the division of BAs detected by cluster analysis (Figure 7). The first two correspondence axes explain roughly 67% of the variability in the dataset. Axis 1 largely defines two major types of assemblages (LQ vs UQ) and likely represents stratigraphy (temporal variation). Ptychagnostus and Spartobranchus are strongly associated with the LQ BAs. Axis 2 further discriminates between the UQ1 and UQ2 BAs. Taxa that are highly recurrent throughout the quarry are found close to the centre of this axis, such as

Yawunik, Sidneyia and Oesia (Figure 7). Taxa that are abundant but have more stratigraphically

163 structured distributions such as Haplophrentis and Kootenayscolex are plotted further from the centre of the ordination (closer to UQ1 and UQ2, respectively).

3.2 Patterns of ecological change at Marble Canyon

From BA 230 to BA 350 and in BA 370, epibenthic-sessile-suspension feeding (ESSU) is the most common ecological mode, generally constituting over 50% of the entire ecological mode occupancy (except in the case of BA 370, where it constitutes 47% of the assemblage ecological mode occupancy; Figure 8). In BA 360 as well as BA 380 to BA 420, nektonic- vagrant-hunting/scavenging (NKHS) is the most common ecological mode. In BA 480 to BA

510, nektobenthic-vagrant-deposit feeding (NKDE) is the most common ecological mode.

Together, these three ecological mode (ESSU, NKHS, NKDE) constitute at least 72% the total ecological mode occupancy of their respective BA (BA 510) and at most 97% (BA 500).

The results of clustering analysis using Morisita-Horn recapitulate the BA-level quantitative patterns of dominance described above (Figure 9). Ecological group 1 (EG1) consists of the BAs dominated by the ESSU ecological mode; ecological group 2 (EG2) consists of the BAs dominated by the NKHS ecological mode; and ecological group 3 (EG3) consists of the BAs dominated by the NKDE ecological mode (Figure 9). These clusters were used to build the convex hulls in the correspondence analysis of the Marble Canyon BA-level ecological compositions (Figure 10). While the ESSU, NKHS, and NKDE ecological mode are all highly abundant at Marble Canyon, they do not plot toward the origin of the ordination due to their strong associations with particular strata.

164

3.2 Burgess Shale multi-locality patterns

Bedding assemblage-level patterns of change in both the Marble Canyon and Walcott

Quarry show a balance between periods of relative faunal stability and rapid species turnover

(Figures 5, 6). At Walcott Quarry, this was taken as an indication of periodic environmental disturbance followed by rapid recolonization (Caron and Jackson 2008), often resulting in a very similar fauna in successive bedding assemblages. This pattern is even more pronounced at

Marble Canyon, where long periods of compositional stasis are only rarely interrupted by a rapid change in species assemblage, which is in turn followed by a period of compositional stasis

(Figure 5,6). Only two major turnover events occur at Marble Canyon after which there is a significant change in faunal composition: the change from the UQ1 fauna to UQ2 fauna

(approximately between BA 340 and BA 350) and the change from the UQ2 fauna to the LQ fauna (after BA 420). The patterns of shifting ecological modes from the Marble Canyon bedding assemblages also provide some support for this view. The clustering of ecological groups and taxonomic groups closely mirror each other, suggesting that the factors that caused major changes in faunal composition similarly caused a change in the environmental conditions of Marble Canyon, favouring a new faunal assemblage with a new ecological structure than what had existed previously.

The most significant differences between the abundance-based and presence/absence taxonomic cluster analyses of the Burgess Shale dataset including all four sampled localities

(Walcott Quarry [WQ], Raymond Quarry [RQ], Tulip Beds [TB] and Marble Canyon [MC]) are the positioning of the TB, and whether or not the WQ forms a single cluster (Figure 11). Using

Jaccard (Figure 11B), all major clusters are reciprocally exclusive of BAs from different localities. Using Morisita-Horn, the UQ1 BAs from MC cluster with a subset of the WQ sites

165 and the LQ BAs cluster with the TB (Figure 11A). The clustering of UQ1 MC and some WQ sites is likely due to the great abundance of Liangshanella in these BAs; the clustering of LQ

BAs and TB is explained by the presence of Peronopsis, a statistically significant indicator species for an MC+TB-type bedding assemblage (Table 1).

Roughly 67% of the total variance in the Burgess Shale taxonomic dataset is represented by the first two axes of our detrended correspondence analysis (Figure 12). Four non-overlapping groups of BAs represent MC, WQ, RQ and TB (Figure 12). These groups are easily distinguished on Axis 1, which explains 39% of the variability in the dataset and suggests that the WQ BAs are most representative of the Burgess Shale fauna overall, as they cluster near the centre of the ordination. On Axis 1, MC and TB are most distant from each other, suggesting these faunas to be the most highly differentiated within our dataset. While this does support the

TB and MC localities being the most stratigraphically disparate pair of localities based on species composition, the RQ BAs are to the right of the WQ BAs, such that reading the ordination from left to right does not precisely re-capitulate our inferred stratigraphic ordering of sites from young to old. This suggests that Axis 1 does not directly represent time, but rather an environmental variable co-varying with time that changes slowly enough that it was not significantly different between the two temporally intermediate localities (WQ and RQ). Within this framework, we interpret the stratigraphic order of the included localities to be (from youngest to oldest): MC, RQ, WQ, TB. This based on multiple lines of evidence: i) The TB is part of the Campsite Cliff Shale Member of the Stephen Formation, which is stratigraphically below the Walcott Quarry Member, which includes both the Walcott and

Raymond quarries (Fletcher and Collins 1998; O’Brien and Caron 2015). ii) The Stephen-Eldon contact point lies directly above MC (Caron et al. 2014).

166 iii) MC is part of the Ehmaniella biozone, within the uppermost part of the Cambrian Stage 5

(Caron et al. 2014). iii) Our correspondence analysis of the species abundance patterns of the four best sampled

Burgess Shale localities suggests that the MC and TB are the most compositionally differentiated localities. The WQ and RQ faunas are found to be intermediate to MC and TB along the correspondence axis which explains the great amount of variability within the ecological dataset, suggesting that these faunas to be transitional in their composition. iv) Our indicator species analysis identifies 21 taxa (Table 1: RQ+WQ, WQ+TB, MC+RQ+WQ, and RQ+WQ+TB) that are congruous with the relative ordering of the studied localities to be

MC, RQ, WQ, TB. Only 10 taxa that support a different relationship between localities (Table 1:

MC+TB, RQ+TB).

While these data do not allow us to completely disregard the possibility that discontinuous sedimentation between MC and the other localities obscure its actual stratigraphic position, further information (ie. geochemical data, more comparable localities, a wider sampled section of RQ) would be required to refine our hypothesis further.

Statistically significant indicator species are plotted onto the CA of the Burgess Shale by species composition in the colour corresponding to their locality; species names in black are indicators for a combination of more than one locality (Table 1). Each locality contains species that act as strong indicators for their BA compositions and which do not seem to share any taxonomic affinity with each other; arthropods, hemichordates, annelids, sponges, algae, and taxa of unknown affinity are all included among the locality-specific indicator species (Figure 12;

Table 1).

167

Ecologically, four major clusters describe the BAs of our four-locality Burgess Shale dataset (Figure 13), however, each includes stratigraphically disparate BAs. These clusters can be defined by the most dominant trophic mode present: the largest cluster represents suspension feeding-dominated BAs and contains BAs from all four localities, including the TB bulk assemblage; the second-largest cluster represents BAs dominated by hunter/scavenger taxa and includes BAs from WQ, RQ and MC; the last two smaller clusters are dominated by deposit feeders and include BAs from WQ and MC (specifically, the LQ BAs from MC). When used to plot convex hulls on the correspondence analysis of the Burgess Shale ecological matrix, there is minimal overlap among these clusters (Figure 14). The first two axes of this analysis explain over 52% of the total variation in the community dataset, suggesting that the relative proportions of represented trophic modes is of principal importance when defining ecological structure among BAs. Despite recurring through all included localities, the ESSU, NKHS, and NKDE niches are offset from the centre of the ordination, corresponding to the subset of BAs in which they are dominant (Figure 14).

4. DISCUSSION

4.1 The short timescale structuring of the Burgess Shale community

Broadly, we can expect the compositional change of ecological communities to conform to one of two sets of patterns. The first is that species or individuals are distributed entirely randomly; this view can be described as the null model of community assembly, with the unified neutral theory of biodiversity being its most influential incarnation (Hubbell 2005; Rosindell et al. 2011). Alternatively, community composition and change may be structured by a combination of both abiotic and biotic factors, resulting in non-random patterns of species distributions over

168 time. There is still considerable debate regarding the relative importance of these factors on community assembly, part of which is due to the temporal constraints of neontological datasets.

This is particularly true of experimental studies analyzing trends in diversity change (Simberloff and Wilson 1969; Findlay and Kasian 1986; Condit et al. 1999), but even observational studies rarely span more than a few decades in length (Grant and Grant 2002; Azaele et al. 2006).

Paleontological datasets are unique in this regard, providing a much broader temporal sampling than is otherwise possible.

Our data suggests that even at the level of 10cm thick bedding assemblages, an uncommonly short timescale for a paleontological dataset, Cambrian communities alternated between periods of relative faunal stability and rapid species turnover (Figures 5, 6). At Walcott

Quarry, this was taken as an indication of periodic environmental disturbance followed by rapid recolonization (Caron and Jackson 2008), often resulting in a very similar fauna between successive bedding assemblages. This pattern is even more pronounced at Marble Canyon, where long periods of compositional stasis are only rarely interrupted by a rapid change in species assemblage, which is in turn followed by a period of compositional stasis (Figure 5,6). Only two major turnover events occur at Marble Canyon after which there is a significant change in faunal composition: the change from the UQ1 fauna to UQ2 fauna and the change from the UQ2 fauna to the LQ fauna. The fact that the clustering of ecological groups and taxonomic groups closely mirror each other further suggests that the factors that caused major changes in faunal composition similarly caused a change in the environmental conditions of Marble Canyon, favouring a new faunal assemblage with a new ecological structure than what had existed previously.

169

Additionally, most bedding assemblages of both quarries are numerically dominated by taxa with broad stratigraphic distributions (Figure 1). At Marble Canyon, these taxa are Oesia,

Haplophrentis and Liangshanella; at Walcott Quarry, Liangshanella, Hazelia, and Marrella are the most temporally recurrent taxa. The dominance of these temporally widespread taxa further suggests that dispersal and colonization ability significantly influenced the short timescale assembly of marine communities during the above described periods of compositional stasis. The nature of the perturbations which resulted in major compositional change, on the other hand, must remain speculative. However, a variety of abiotic factors are known to result in major regime shifts in modern marine communities (Sousa 1984; Conversi et al. 2015) such as the physical removal of encrusting species (Altman and Whitlatch 2007), changing proportions of different sediment types (McArthur et al. 2010), and shifting temperatures and perturbations of ocean chemistry (Harley et al. 2006; Bornette and Puijalon 2011). These types of environmental fluctuations affect a myriad of biological factors, such as competition and larval recruitment, which may explain the non-random changes in community composition found at both Marble

Canyon and Walcott Quarry.

4.2 Macroecological patterns of community structure across the Burgess Shale

Our data incorporating four well-sampled Burgess Shale sites (Walcott Quarry, Raymond

Quarry, Tulip Beds and Marble Canyon) reveals major patterns of taxonomic compositional change that are at odds with a random distribution of taxa from a single species pool. One of the most striking examples of this temporal variation in species composition among localities is the abundance of sponges. They are the second-most diverse phylum at the Tulip Beds and Walcott

Quarry, and the second-most abundant phylum at the Walcott Quarry and Raymond Quarry, but

170 at Marble Canyon, only five specimens out of 16,431 are sponges. The priapulids are similarly abundant throughout most well-studied Cambrian localities, particularly at Raymond Quarry, where Ottoia alone makes up 10.1% of specimens from our sampled interval. They are an even more significant part of the older Chengjiang localities, where they are the most abundant taxon at the Shankou site and second-most abundant taxon at Mafang (Zhao et al. 2013; O’Brien and

Caron 2015). At Marble Canyon, however, only a few specimens of Selkirkia and an as-of-yet indeterminate priapulid represent this phylum.

Other taxa show the opposite stratigraphic relationship, systematically increasing in abundance from older to younger localities. While hemichordates are absent from the Tulip

Beds, each of the younger localities has a different hemichordate taxon that is prevalent enough to be an indicator species (Spartobranchus, an undescribed enteropneust, and Oesia for Walcott

Quarry, Raymond Quarry, and Marble Canyon, respectively). The only annelid at Tulip Beds is a single specimen of Burgessochaeta. By contrast, the younger Walcott Quarry and Marble

Canyon possesses significant numbers of annelids (Burgessochaeta and Kootenayscolex, respectively).

A number of possibilities exist to explain the high degree of taxonomic heterogeneity among our sampled localities. The first are ecological interactions mediating species distributions. Inter-specific competition may have sequentially excluded species from being present in younger localities. For example, sponges may have been systematically outcompeted in the epifaunal suspension feeding niche by taxa such as Haplophrentis (Moysiuk et al. 2017) and Oesia (Hardin 1960). The near-complete absence of Burgessochaeta may be due to its niche being filled Kootenayscolex, which is thought to share a similar mode of life (Caron and Jackson

2008; Nanglu and Caron 2018). The same could be argued of other taxa which presumably share

171 a significant degree of ecological overlap and show a similar pattern of distributional turnover, such as Yawunik (Aria et al. 2015) and Tokummia (Aria and Caron 2017) at Marble Canyon taking the place of Leanchoilia and Branchiocaris from older localities.

Biotic drivers of community assembly, while intuitively interesting due to their parallels suggested mechanisms in modern environments such as competitive exclusion (Connell 1961a, b; McCook et al. 2001), are difficult to prove in a paleontological setting. First, the exact niches of these taxa, while perhaps broadly similar, may have been sufficiently differentiated such that competitive exclusion may not have been a major factor in their distributions (MacArthur 1958;

Hardin 1960; Levine and HilleRisLambers 2009). Second, the extent to which competitive factors actually structure communities is still an intensely debated topic, with much of the current consensus being that dispersal ability is at least an equally important factor (Rosindell et al. 2011; Klompmaker and Finnegan 2018). Third, competitive exclusion would require direct interaction, but non-contiguous sampling of strata between localities precludes a more nuanced view of the gradual shifts in taxonomic composition that would be required to corroborate a model of ecological-replacement driven turnover.

The other possibility for explaining the significant local variation in community composition are gradually shifting abiotic factors. For example, the near total absence of sponges from Marble Canyon could be due to either turbid conditions impairing their ability to feed

(Pineda et al. 2017) or by a lack of available substrate for larval recruitment (Maldonado and

Young 1996; Whalan et al. 2015). Within this context, the abiotic drivers of community change described above (4.1), which may have caused only subtle fluctuations over short time scales, may have acted over the course of the entire Burgess Shale to produce significantly different environments and faunas. While this remains the most plausible cause for the observed changes

172 in species composition, our conclusions must remain speculative until they can be corroborated by geochemical or paleoenvironmental data.

Our view of Cambrian trophic ecology is also impacted through our quantitative BA- level analysis of multiple localities. The view that Cambrian paleocommunities were relatively stable in their ecological structure has been largely based on limited comparisons between sites at the level of broadly sampled bulk assemblages without quantitative comparisons of how within-locality ecological fluctuations compared across sites (Zhao et al. 2013; O’Brien and

Caron 2015). For example, a bulk assemblage comparison of the Tulip Beds and Walcott Quarry indicated that there was a broadly similar diversity of ecological strategies, and that this pattern was also shared with the Chengjiang localities (O’Brien and Caron 2015). Our data suggests that if we consider fine scale stratigraphic variation rather than combining the ecologies of each major locality into a single pool, the dominant ecological modes were susceptible to considerable variation.

This is most starkly demonstrated by the ecological positioning of the Tulip Beds next to the Upper Quarry 1 sites from Marble Canyon (Figures 13, 14). Within our dataset, these strata represent the oldest and youngest sampled units, respectively. However, they are more similar in their community structure (as viewed through niche occupation) than they are to any intervening bedding assemblages. Similarly, the arthropod-rich Upper Quarry 2 bedding assemblages from

Marble Canyon are more ecologically similar to bedding assemblages from the Walcott Quarry and Raymond Quarry than they are to the Marble Canyon assemblages directly above and below them (Figure 13, 14). These results demonstrate that at the BA-level view of ecological structure in the Burgess Shale, there are three broad types of fauna typified by which trophic mode is dominant: macrofaunal heterotrophy (hunters and scavengers), sessile suspension feeding, or

173 deposit feeding (Figures 13, 14). Further, the bedding assemblages that fall into these categories do not show a strict unidirectional temporal trend. Rather, bedding assemblages may be more similar in ecological structure to other, temporally disjunct bedding assemblages than to their stratigraphically adjacent neighbours. In this context, ecological structure throughout the Burgess

Shale was highly heterogeneous both at the level of bedding assemblages and localities.

These conclusions also require a reappraisal of the role of predation in mediating

Cambrian community ecology. Predation is often suggested as one of the principal forces structuring communities during the Cambrian (Bengtson 2002; Hu et al. 2007; Vannier et al.

2007; Vannier 2012; Laflamme et al. 2013), as well as one of the factors most distinguishing

Cambrian communities from the preceding Ediacaran fauna (Laflamme and Narbonne 2008;

Xiao and Laflamme 2009). Our data is somewhat at odds with this view; the epifaunal suspension feeding cluster is the largest cluster of sites in our dataset, suggesting that this mode of life was at least as important as predation for defining community ecological structure. One way of interpreting this data would be to suggest periodic alternation between top-down

(predator and competition-mediated) vs. bottom-up (prey and environment-mediated) control of community ecology (Roff et al. 2016). However, it has been noted that this dichotomy may not be appropriate in marine settings, as these mechanisms are not mutually exclusive (Conversi et al. 2014). For example, many predatory species have larval forms (which we are unlikely to see in the fossil record, except in the most exceptional of circumstances (Maas et al. 2006; Yin et al.

2007)) which are highly sensitive to environmental conditions, while sessile organisms are frequently in direct competition for substrate (McCook et al. 2001). It is, therefore, most appropriate to view the structure of Cambrian paleocommunities as we now view modern benthic communities: the result of multiple inter-specific drivers, such as predation (Connell

174

1961b), competition (Connell 1961a; McCook et al. 2001) and niche partitioning (MacArthur

1958), acting simultaneously on the backdrop of abiotic factors such as environmental perturbation (Krebs 2009; Conversi et al. 2014; Pershing et al. 2015).

5. CONCLUSIONS

While our understanding of the Cambrian Explosion continues to progress, the discovery of the Marble Canyon fossil site clearly demonstrates that major discoveries at the level of entire faunas remain to be made. Such discoveries have the potential not only to shed light on the early origins of nearly all major metazoan phyla, but also on the structure and diversity of Cambrian ecosystems. Our study represents the first multivariate, quantitative analysis of the four best- sampled Burgess Shale localities and the largest analysis of Cambrian communities to date. The integration of fine resolution stratigraphic data possesses significant promise for disentangling the potential mechanisms behind the assembly of some of the earliest complex animal communities. Bedding assemblage-level changes in both taxonomic and ecological composition show that periods of relative stasis at short-time scales are common among Cambrian localities.

Widening the scope of our observations to stratigraphically disparate localities reveals that the dominance of different ecological modes was patchy throughout the Burgess Shale, and that major taxonomic changes occurred between localities. These data indicate that the Burgess Shale was highly heterogeneous with respect to both ecology and species composition.

175

ACKNOWLEDGEMENTS

We thank P. Fenton and M. Akrami for collections assistance at the Royal Ontario Museum. We thank T. M. Cullen for feedback and M. Orobko for assistance with R coding. We also thank R.R

Gaines for assistance with calibrating Marble Canyon strata and feedback regarding the depositional environment. Material for this study was collected under several Parks Canada

Research and Collections permits (to J.-B. Caron). Major funding support for field work comes from the Royal Ontario Museum (Research and collection grants), the National Geographic

Society (2014 research grant to J.-B. Caron), the National Science Foundation (2016 EAR-

1556226 Award to Robert Gaines-Pomona College). K. Nanglu’s doctoral research is supported by fellowships from the University of Toronto (Department of Ecology and Evolutionary

Biology) and J.-B. Caron’s NSERC Discovery Grant (number 341944). This is Royal Ontario

Museum Burgess Shale project number 81.

176

REFERENCES

Altman, S., and R. B. Whitlatch. 2007: Effects of small-scale disturbance on invasion success in

marine communities. Journal of Experimental Marine Biology and Ecology 342:15–29.

Aria, C., and J. B. Caron. 2015: Cephalic and limb anatomy of a new isoxyid from the burgess

shale and the role of “Stem bivalved arthropods” in the disparity of the frontalmost

appendage. PLoS ONE 10:1–37.

Aria, C., and J.-B. Caron. 2017: Burgess Shale fossils illustrate the origin of the mandibulate

body plan. Nature 545:89–92.

Aria, C., J.-B. Caron, and R. Gaines. 2015: A large new leanchoiliid from the Burgess Shale and

the influence of inapplicable states on stem arthropod phylogeny. Palaeontology 58:629–

660.

Azaele, S., S. Pigolotti, J. R. Banavar, and A. Maritan. 2006: Dynamical evolution of

ecosystems. Nature 444:926–928.

Bambach, R. K., A. M. Bush, and D. H. Erwin. 2007: Autecology and the filling of ecospace:

key metazoan radiations. Palaeontology 50:1–22.

Barwell, L. J., N. J. B. Isaac, and W. E. Kunin. 2015: Measuring species-diversity with species

abundance data. Journal of Animal Ecology 84:1112–1122.

Beli, E., S. Piraino, and C. B. Cameron. 2017: Fossilization processes of graptolites: insights

from the experimental decay of Rhabdopleura sp. (Pterobranchia). Palaeontology 60:389–

400.

Bengtson, S. 2002: Origins and early evolution of predation. Paleontological Society 8:289–318.

Bornette, G., and S. Puijalon. 2011: Response of aquatic plants to abiotic factors: A review.

Aquatic Sciences 73:1–14.

177

De Cáceres, M., P. Legendre, S. K. Wiser, and L. Brotons. 2012: Using species combinations in

indicator value analyses. Methods in Ecology and Evolution 3:973–982.

Caron, J.-B., and D. A. Jackson. 2006: Taphonomy of the Greater Phyllopod Bed Community,

Burgess Shale. Palaios 21:451–465.

Caron, J.-B., and D. A. Jackson. 2008: Paleoecology of the Greater Phyllopod Bed community,

Burgess Shale. Palaeogeography, Palaeoclimatology, Palaeoecology 258:222–256.

Caron, J.-B., R. R. Gaines, C. Aria, M. G. Mángano, and M. Streng. 2014: A new phyllopod bed-

like assemblage from the Burgess Shale of the Canadian Rockies. Nature communications

5:3210.

Colwell, R. K., A. Chao, N. J. Gotelli, S. Y. Lin, C. X. Mao, R. L. Chazdon, and J. T. Longino.

2012: Models and estimators linking individual-based and sample-based rarefaction,

extrapolation and comparison of assemblages. Journal of Plant Ecology 5:3–21.

Condit, R., P. S. Ashton, N. Manokaran, J. V LaFrankie, S. P. Hubbell, and R. B. Foster. 1999:

Dynamics of the forest communities at Pasoh and Barro Colorado : Comparing two 50-ha

plots. Philosophical Transactions of the Royal Society B: Biological Sciences 354:1739–

1748.

Connell, J. H. 1961a: The influence of interspecific competition and other factors on the

distribution of the barnacle Chthamalus Stellatus. Ecology 42:710–723.

———. 1961b: Effects of competition, predation by Thais lapillus, and other factors on natural

populations of the barnacle Balanus balanoides. Ecological Monographs 31:61–104.

Conversi, A., V. Dakos, A. Gardmark, S. Ling, C. Folke, P. J. Mumby, C. Greene, M. Edwards,

T. Blenckner, M. Casini, A. Pershing, and C. Mollmann. 2014: A holistic view of marine

regime shifts. Philosophical Transactions of the Royal Society B: Biological Sciences 370:

178

1-8.

Conway Morris, S. 1986: The community structure of the Middle Cambrian phyllopod bed

(Burgess Shale). Palaeontology 29:423–467.

Devereux, M. G. 2001: Palaeoecology of the Middle Cambrian Raymond Quarry fauna, Burgess

Shale, British Columbia. University of Western Ontario Master’s Thesis.

Erwin, D. H., M. Laflamme, S. M. Tweedt, E. A. Sperling, D. Pisani, and K. J. Peterson. 2011:

The Cambrian conundrum: early divergence and later ecological success in the early history

of animals. Science 334:1091–1097.

Findlay, D. L., and S. E. M. Kasian. 1986: Phytoplankton community responses to acidification

of lake 223, experimental lakes area, northwestern Ontario. Water, Air, and Soil Pollution

30:719–726.

Fletcher, T., and D. Collins. 1998: The Middle Cambrian Burgess Shale and its relationship to

the Stephen Formation in the southern Canadian Rocky Mountains. Canadian Journal of

Earth Sciences 35:413–436.

Fletcher, T. P., and D. H. Collins. 2003: The Burgess Shale and associated Cambrian formations

west of the Fossil Gully Fault Zone on Mount Stephen, British Columbia. Canadian Journal

of Earth Sciences 40:1823–1838.

Gaines, R., E. Hammarlund, X. Hou, C. Qi, S. Gabbott, Y. Zhao, J. Peng, and D. Canfield. 2012:

Mechanism for Burgess Shale-type preservation.:1–5.

Gaines, R. R., and M. L. Droser. 2010: The paleoredox setting of Burgess Shale-type deposits.

Palaeogeography, Palaeoclimatology, Palaeoecology 297:649–661.

Gaines, R. R., D. E. G. Briggs, and Y. Zhao. 2008: Cambrian Burgess Shale-type deposits share

a common mode of fossilization. Geology 36:755–758.

179

Grant, P. R., and B. R. Grant. 2002: Unpredictable Evolution in a 30-Year Study of Darwin’s

Finches. Science 296:707–711.

Hardin, G. 1960: The Competitive Exclusion Principle. Science 131:1292–1297.

Harley, C. D. G., A. R. Hughes, K. M. Hultgren, B. G. Miner, C. J. B. Sorte, C. S. Thornber, L.

F. Rodriguez, L. Tomanek, and S. L. Williams. 2006: The impacts of climate change in

coastal marine systems. Ecology Letters 9:228–241.

Hu, S., M. Steiner, M. Zhu, B. D. Erdtmann, H. Luo, L. Chen, and B. Weber. 2007: Diverse

pelagic predators from the Chengjiang Lagerstätte and the establishment of modern-style

pelagic ecosystems in the early Cambrian. Palaeogeography, Palaeoclimatology,

Palaeoecology 254:307–316.

Hubbell, S. P. 2005: The neutral theory of biodiversity and biogeography and Stephen Jay

Gould. Paleobiology 31:122–132.

Klompmaker, A. A., and S. Finnegan. 2018: Extreme rarity of competitive exclusion in modern

and fossil marine benthic ecosystems. Geology 46:723–726.

Koleff, P., P. Koleff, K. J. Gaston, K. J. Gaston, J. J. Lennon, and J. J. Lennon. 2003: Measuring

beta diversity for presence –absence data. Journal of Animal Ecology 72:367–382.

Krebs, C. J. 2009: Ecology. Pearson Education, Inc., San Francisco, California.

Laflamme, M., and G. M. Narbonne. 2008: Competition in a Precambrian world : palaeoecology

of Ediacaran fronds Leaf-shaped Ediacaran fronds are among the most widespread and

readily. Geology 24:182–187.

Laflamme, M., S. A. F. Darroch, S. M. Tweedt, K. J. Peterson, and D. H. Erwin. 2013: The end

of the Ediacara biota: Extinction, biotic replacement, or Cheshire Cat? Gondwana Research

23:558–573.

180

Levine, J. M., and J. HilleRisLambers. 2009: The importance of niches for the maintenance of

species diversity. Nature 461:254–257.

Maas, A., A. Braun, X. P. Dong, P. C. J. Donoghue, K. J. Müller, E. Olempska, J. E. Repetski,

D. J. Siveter, M. Stein, and D. Waloszek. 2006: The ’Orsten’-More than a Cambrian

Konservat-Lagerstätte yielding exceptional preservation. Palaeoworld 15:266–282.

MacArthur, R. H. 1958: Population ecology of some warblers of Northeastern coniferous forests.

Ecology 39:599–619.

Magurran, A. E. 2004: Measuring Biological Diversity. Blackwell Science Ltd., Oxford.

Maldonado, M., and C. M. Young. 1996: Effects of physical factors on larval behavior,

settlement and recruitment of four tropical demosponges. Marine Ecology Progress Series

138:169–180.

McArthur, M. A., B. P. Brooke, R. Przeslawski, D. A. Ryan, V. L. Lucieer, S. Nichol, A. W.

McCallum, C. Mellin, I. D. Cresswell, and L. C. Radke. 2010: On the use of abiotic

surrogates to describe marine benthic biodiversity. Estuarine, Coastal and Shelf Science

88:21–32.

McCook, L. J., J. Jompa, and G. Diaz-Pulido. 2001: Competition between corals and algae on

coral reefs: A review of evidence and mechanisms. Coral Reefs 19:400–417.

Morris, S. C., and J.-B. Caron. 2014: A primitive fish from the Cambrian of North America.

Nature.

Moysiuk, J., M. Smith, and J.-B. Caron. 2017: Hyoliths are Palaeozoic lophophorates. Nature

541:394–397.

Nanglu, K., and J.-B. Caron. 2018: A new Burgess Shale polychaete and the origin of the annelid

head revisited. Current Biology 28:319–326.

181

Nanglu, K., J.-B. Caron, and C. B. Cameron. 2015: Using experimental decay of modern forms

to reconstruct the early evolution and morphology of fossil enteropneusts. Paleobiology 41:

460-478.

Nanglu, K., J.-B. Caron, S. Conway Morris, and C. B. Cameron. 2016: Cambrian suspension-

feeding tubicolous hemichordates. BMC Biology 14:9.

O’Brien, L. J., and J.-B. Caron. 2012: A new stalked Filter-Feeder from the Middle Cambrian

Burgess Shale, British Columbia, Canada. PLoS ONE 7.

O’Brien, L. J., and J.-B. Caron. 2015: Paleocommunity Analysis of the Burgess Shale Tulip

Beds , Mount Stephen , British Columbia : Comparison with the Walcott Quarry and

Implications for Community Variation in the Burgess Shale. Paleobiology:1–27.

O’Brien, L. J., J.-B. Caron, and R. R. Gaines. 2014: Taphonomy and depositional setting of the

Burgess Shale Tulip Beds, Mount Stephen, British Columbia. PALAOIS 29:309–324.

Pershing, A. J., K. E. Mills, N. R. Record, K. Stamieszkin, K. V. Wurtzell, C. J. Byron, D.

Fitzpatrick, W. J. Golet, and E. Koob. 2015: Evaluating trophic cascades as drivers of

regime shifts in different ocean ecosystems. Philosophical Transactions of the Royal

Society B: Biological Sciences 370:1–9.

Pineda, M. C., B. Strehlow, M. Sternel, A. Duckworth, J. Den Haan, R. Jones, and N. S.

Webster. 2017: Effects of sediment smothering on the sponge holobiont with implications

for dredging management. Scientific Reports 7:1–15.

Roff, G., C. Doropoulos, A. Rogers, Y. M. Bozec, N. C. Krueck, E. Aurellado, M. Priest, C.

Birrell, and P. J. Mumby. 2016: The ecological role of sharks on coral reefs. Trends in

Ecology and Evolution 31:395–407.

Rosindell, J., S. P. Hubbell, and R. S. Etienne. 2011: The unified neutral theory of biodiversity

182

and biogeography at age ten. Trends in Ecology & Evolution 26:340–348.

Simberloff, D. S., and E. O. Wilson. 1969: Experimental zoogeography of islands: the

colonization of empty islands. Ecology 50:278–296.

Sousa, W. W. 1984: The role of disturbance in natural communities. Annual review of ecology

and Systematics 15:353–391.

Vannier, J. 2012: Gut contents as direct indicators for trophic relationships in the Cambrian

marine ecosystem. PLoS ONE 7.

Vannier, J., M. Steiner, E. Renvoisé, S.-X. Hu, and J.-P. Casanova. 2007: Early Cambrian origin

of modern food webs: evidence from predator arrow worms. Proceedings of the Royal

Society Biological sciences 274:627–633.

Whalan, S., M. A. Abdul Wahab, S. Sprungala, A. J. Poole, and R. De Nys. 2015: Larval

settlement: The role of surface topography for sessile coral reef invertebrates. PLoS ONE

10:1–17.

Xiao, S., and M. Laflamme. 2009: On the eve of animal radiation: phylogeny, ecology and

evolution of the Ediacara biota. Trends in Ecology and Evolution 24:31–40.

Yin, L., M. Zhu, A. H. Knoll, X. Yuan, J. Zhang, and J. Hu. 2007: Doushantuo embryos

preserved inside diapause egg cysts. Nature 446:661–663.

Zhao, F., J.-B. Caron, D. J. Bottjer, S. Hu, Z. Yin, and M. Zhu. 2013: Diversity and species

abundance patterns of the early Cambrian (Series 2, Stage 3) Chengjiang Biota from China.

Paleobiology 40:50–69.

183

Tables

Localities Indicator taxon A-component B-component Indicator stat p-value Marble Canyon Oesia disjuncta 0.9980 1.0000 0.999 0.005 (MC) New arthropod J 1.0000 0.9444 0.972 0.005 Indet worm A 1.0000 0.8333 0.913 0.015 Tokummia 1.0000 0.8333 0.913 0.020 Metaspriggina 0.9736 0.8333 0.901 0.020 Yawunik 1.0000 0.7778 0.882 0.020 Cyanobacteria-MC 1.0000 0.6667 0.816 0.045 Ptychoparid-MC 0.9909 0.6667 0.813 0.035 Raymond Quar Herpetogaster collinsi 0.9536 1.0000 0.977 0.005 ry (RQ) Allonia 1.0000 0.9333 0.966 0.005 Fuxianospira 0.9365 0.9333 0.935 0.005 Medusoid_RQ 1.0000 0.8667 0.931 0.020 Helcionellid 1.0000 0.8000 0.894 0.035 Enteropneust RQ-A 1.0000 0.8000 0.894 0.045 Naraoia magna 1.0000 0.2000 0.447 0.035 Walcott Quarry Burgessia bella 1.0000 0.9583 0.979 0.005 (WQ) Ptychagnostus praecurrens 0.9624 0.9583 0.960 0.005 Morania sp. 1.0000 0.9167 0.957 0.005 Spartobranchus tenuis 0.9588 0.8333 0.894 0.035 Tulip Beds (T Ogygopsis klotzi 1.0000 1.0000 1.0000 0.035 B) Glossopleura 1.0000 1.0000 1.0000 0.035 Siphusauctum.gregarium 1.0000 1.0000 1.0000 0.035 MC+TB Peronopsis 0.9992 1.0000 1.0000 0.005 RQ+TB Anomalocaris canadensis 0.9994 1.0000 1.000 0.005 Vauxia sp. 0.9925 1.0000 0.996 0.005 Tubullela sp. 0.9861 1.0000 0.993 0.005 Leanchoilia spp. 0.9783 1.0000 0.989 0.005 Hurdia sp. 0.9455 1.0000 0.972 0.005 Micromitra burgessensis 0.8999 1.0000 0.949 0.005 Choia spp. 0.9798 0.8750 0.926 0.045 Mackenzia costalis 0.9664 0.8750 0.920 0.005 Tuzoia spp. 0.8156 1.0000 0.903 0.005 RQ+WQ Pagetia bootes 1.0000 0.8718 0.934 0.005 Pollingeria grandis 1.0000 0.8718 0.934 0.005 Olenoides serratus 1.0000 0.8205 0.906 0.005 WQ+TB Wiwaxia corrugata 1.0000 1.0000 1.000 0.005 Hazelia spp. 0.9964 1.0000 0.998 0.005 Selkirkia spp. 0.9928 1.0000 0.996 0.005 Ehmaniella spp. 0.9889 1.0000 0.994 0.005 Marrella splendens 1.0000 0.9600 0.980 0.005 Scenella amii 0.9963 0.9600 0.978 0.005 Yohoia tenuis 1.0000 0.9200 0.959 0.005 Canadaspis perfecta 0.9941 0.8800 0.935 0.010 Laggania cambria 1.0000 0.8400 0.917 0.025 Naraoia spp. 0.9922 0.8400 0.913 0.030 Lingulella waptaensis 0.9762 0.8400 0.906 0.025 Takakkawia lineata 0.9535 0.7600 0.851 0.035 Kootenia burgessensis 0.9288 0.7600 0.840 0.030 MC+RQ+WQ Liangshanella spp. 1.0000 1.0000 1.0000 0.035 RQ+WQ+TB Waptia fieldensis 1.0000 0.9500 0.975 0.005 Diraphora bellicostata 1.0000 0.9250 0.962 0.005 Ottoia prolifica 1.0000 0.9250 0.962 0.005 Isoxys spp. 0.9791 0.8750 0.926 0.005

184

Table 1. Species indicator analysis for the Burgess Shale. The indicator statistic has two components. The A-component describes the likelihood of sampling from the indicated locality

185

Figures

186

Figure 1. Stratigraphic column of the Marble Canyon quarry and broad diversity trends. A: The top 10 taxa are a mix of different taxonomic groups, but are dominated by arthropods. The first four most abundant taxa are stratigraphically widespread throughout the quarry; the next 6 are more restricted (with the exception of Sidneyia). B: Major taxonomic groups represented at

Marble Canyon by abundance and diversity. Arthropods dominate both categories, but lophophorates, hemichordates, and annelids are significant components of the community.

Acronyms: ALGA – algae, ANNE – Annelida, ARTH – Arthropoda, CHORD – Chordata,

187

188

Figure 2. Rarefaction curves for the four studied Burgess Shale localities at the level of bulk assemblages, and the individual Marble Canyon Quarry bedding assemblages. A: At the bulk assemblage level, rarefaction curves have generally plateaued, suggesting that they have been sampled sufficiently for diversity comparisons. B: Rarefaction curves for the Marble Canyon bedding assemblages, with extrapolated curves in dashed lines. For clarity, the 95% confidence intervals are only plotted for BA 350, BA 380, and BA 300. While the rank order of species richness changes slightly with extrapolated curves, the bounds of the confidence intervals make comparisons difficult to validate between any BAs except for the richest (BAs 340 and 350) and poorest (BAs 300 and 370).

189

190

Figure 3. Observed, inferred, and extrapolated species richnesses across BAs with 95% confidence intervals. Only BAs with greater than 299 specimens were included in quantitative analyses (BAs with fewer than 299 specimens are greyed out in the stratigraphic column to the right). Interpolating species richnesses to 299 specimens erases most patterns of fluctuating diversity. Extrapolating richness to 2,566 specimens suggests the same major trend as the observed pattern, but most bedding assemblages will fall within the confidence interval of the adjacent bedding assemblage.

191

192

Figure 4. Rank abundance curves (also known as Whittaker plots) for the three species-richest and -poorest BAs. A lower slope, as seen in BAs 380, 400, and 350, indicates a greater evenness in abundance among constituent species.

193

194

Figure 5. Proportions of major taxonomic groups among the Marble Canyon bedding assemblages. Lophophorates and hemichordates are the most abundant taxa in the upper levels of the quarry. After BA 340, arthropods become the most numerically dominant major taxon, which persists to the lowest strata of the quarry. Acronyms: ALGA – algae, ANNE – Annelida, ARTH

– Arthropoda, CHORD – Chordata, CTEN – Ctenophora, DINO – Dinocarridida, HEMI –

Hemichordata, INDET – Indeterminate taxon, LOBO – Lobopodia, LOPH – Lophophorata,

MOLL – Mollusca, OTHER – All other taxa not listed, PORI – Porifera, PRIA – Priapulida.

195

196

Figure 6. Taxonomic cluster analyses for the Marble Canyon Quarry using both the Morisita-

Horn and Jaccard indices. Both indices support the same broad clustering topology, with the

Upper Quarry (UQ) and Lower Quarry (LQ) BAs forming mutually exclusive clusters. The only significant difference is the shift in the location of BA 360 between the two major UQ clusters.

197

198

Figure 7. Correspondence analysis for the Marble Canyon Quarry. Axis 1 delineates the lower quarry from the upper quarry and Axis 2 delineates the two sub-divisions within the upper quarry: the highest strata dominated by Oesia and Haplophrentis and the lower strata of the upper quarry dominated by Liangshanella and other arthropods.

199

200

Figure 8. Proportions of major ecological groups among the Marble Canyon bedding assemblages. The upper levels of the quarry are largely composed of epibenthic-sessile- suspension feeders. As arthropods become more abundant, nektonic-vagrant-hunter/scavengers become the dominant niche. The lowest four BAs in the quarry are dominated by epibenthic- vagrant-deposit feeders. Acronyms: EPP – epibenthic primary producer, ESSU – epibenthic- sessile-suspension feeder, EVDE – epibenthic-vagrant-deposit feeder, EVHS – epibenthic vagrant hunter/scavenger, NKSU – nektobenthic vagrant suspension feeder, EVGR – epibenthic vagrant grazer, IVHS – infaunal vagrant hunter scavenger, NKDE – nektobenthic deposit feeder,

NKHS – nektobenthic hunter scavenger, NKSU – nektobenthic suspension feeder, PEHS – pelagic hunter scavenger, PESU – pelagic suspension feeder.

201

202

Figure 9. Cluster analysis for the Marble Canyon ecological abundance matrix using the

Morisita-Horn index. Three major clusters are recovered, corresponding to the uppermost levels of the Upper Quarry (UQ1), the lower levels of the Upper Quarry (UQ2), and the Lower Quarry

(LQ).

203

204

Figure 10. Ecological correspondence analysis for the Marble Canyon Quarry. There are three main types of ecological groups: those dominated by epibenthic-sessile-suspension feeding

(ESSU; purple) those dominated by nektobenthic-vagrant-hunter/scavengers (NKHS; blue) and those dominated nektobenthic-vagrant-deposit feeding (NKDE; pink).

205

206

Figure 11. Taxonomic cluster analyses for the Walcott Quarry (WQ), Raymond Quarry (RQ),

Tulip Beds (TB) and Marble Canyon (MC) using both the Morisita-Horn and Jaccard indices.

207

208

Figure 12. Detrended correspondence analysis for the Walcott Quarry (WQ: red), Raymond

Quarry (RQ: green), Tulip Beds (TB: purple) and Marble Canyon (MC: blue) taxonomic abundance matrix. Indicator species are plotted in the colour of their respective locality.

Indicator species for groups of more than one locality are plotted in black.

209

210

Figure 13. Ecological cluster analyses for the Walcott Quarry (WQ), Raymond Quarry (RQ),

211

212

Figure 14. Ecological correspondence analysis for the Walcott Quarry (WQ), Raymond Quarry

213

Appendices

214

Appendix 1. Cluster analyses of the Marble Canyon bedding assemblages using different thresholds for minimum specimen number/bedding assemblage. To produce cumulative bins, successive adjacent bedding assemblages were summed to reach a minimum specimen number of 300 before being included in the cluster analysis.

215

Acanthotretella Acrothyra Alalcomenaeus Amiskwia Aysheaia strata Ancalagon minor Anomalocaris canadensis spinosa gregaria cambricus sagittiformis pedunculata WQ+120 1 7 0 0 0 0 0 WQ0 1 5 1 0 0 0 0 WQ110 0 2 0 0 0 0 0 WQ120 3 49 581 0 2 0 2 WQ130 12 69 18 0 0 0 8 WQ150 2 13 1 0 0 0 2 WQ180 0 8 1 0 0 0 1 WQ210 0 60 4 1 0 0 3 WQ235 0 6 0 5 0 0 0 WQ245 0 12 0 0 0 0 0 WQ250 0 1 1 0 0 0 1 WQ260 0 5 0 0 0 0 0 WQ265 0 4 0 1 0 0 0 WQ310 0 1 0 0 0 0 0 WQ320 0 0 0 3 0 0 0 WQ370 0 2 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 WQ400 2 2 7 1 0 0 0 WQ420 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 WQ480 0 5 0 2 0 0 0 MC230 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 MC330 0 0 2 0 0 0 0 MC340 1 0 2 0 0 0 0 MC350 0 0 1 0 0 1 0 MC360 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 MC380 0 0 3 0 0 2 0 MC390 0 0 0 0 0 0 0 MC400 0 0 1 0 0 0 0 MC410 1 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 RQ8.0 0 1 0 0 0 10 0 RQ8.2 0 1 0 0 0 15 0 RQ8.3 0 0 0 0 0 18 0 RQ8.4 0 0 1 0 0 15 0 RQ8.5 0 0 0 0 0 8 0 RQ8.6 0 0 1 0 0 9 0 RQ8.8 0 0 0 0 0 16 0 RQ8.9 0 0 0 0 0 7 0 RQ9.0 0 1 0 0 0 19 0 RQ9.1 0 0 0 0 0 9 0 RQ9.2 0 0 0 0 0 7 0 RQ9.3 0 0 0 0 0 9 0 216

RQ9.4 0 0 0 0 0 18 0 RQ9.5 0 0 0 0 0 3 0 RQ9.6 0 0 0 0 0 4 0 TB 0 11 4 0 0 270 0 Branchiocaris Burgessochaeta Cambrorhytium strata Bosworthia spp. Burgessia bella Canadaspis perfecta Canadia setigera pretiosa setigera spp. WQ+120 0 0 4 0 1 2 0 WQ0 1 0 132 14 0 2 12 WQ110 0 0 3 3 0 19 0 WQ120 2 0 136 54 1 37 0 WQ130 0 0 395 5 95 11 5 WQ150 0 0 33 0 31 14 0 WQ180 0 0 13 0 4 10 0 WQ210 0 0 140 0 15 8 0 WQ235 6 0 53 0 4 21 0 WQ245 0 0 19 0 28 0 0 WQ250 2 0 3 0 7 0 0 WQ260 0 0 3 25 0 16 3 WQ265 0 0 2 2 2 13 0 WQ310 0 0 7 0 2 19 1 WQ320 0 0 161 1 0 227 0 WQ370 0 0 0 0 0 60 0 WQ380 0 0 3 9 0 37 0 WQ400 3 0 191 32 0 58 1 WQ420 2 0 1 9 0 6 0 WQ430 0 0 21 0 0 11 0 WQ445 1 0 1 0 0 35 0 WQ455 0 0 12 11 0 4 0 WQ465 0 0 1 0 0 0 0 WQ480 0 3 9 2 0 20 1 MC230 0 0 0 0 0 0 0 MC240 0 0 0 0 0 1 0 MC300 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 MC330 0 0 0 0 0 1 0 MC340 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 MC380 0 0 0 0 0 1 0 MC390 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 MC410 0 0 0 0 0 4 0 MC420 0 0 0 0 0 1 0 MC480 0 0 0 2 0 0 0 MC490 0 0 0 0 0 0 0 MC500 0 0 0 0 0 1 0 RQ8.0 0 0 0 0 1 0 0 RQ8.2 0 0 0 0 1 0 0 RQ8.3 0 1 0 0 0 0 0 RQ8.4 0 0 0 0 0 1 0 RQ8.5 0 0 0 0 1 0 0 RQ8.6 0 1 0 0 2 0 0 RQ8.8 0 0 0 0 0 0 0 RQ8.9 0 0 0 0 0 0 0 217

RQ9.0 0 1 0 1 0 0 0 RQ9.1 0 1 0 0 0 0 0 RQ9.2 0 1 0 0 0 0 0 RQ9.3 0 0 0 0 0 0 0 RQ9.4 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 1 0 RQ9.6 0 0 0 0 0 0 0 TB 4 30 0 1 81 81 0 Capsospongia Caryosyntrips Chancelloria? Chaunograptus Ctenorhabdotu strata Chancia palliseri Choia spp. Crumillospongia spp. undulata serratus spp. scandens s sp. WQ+120 0 0 0 0 0 0 0 0 WQ0 0 0 1 0 0 0 0 2 WQ110 0 0 0 0 0 0 0 0 WQ120 0 0 1 0 0 0 3 0 WQ130 0 0 8 0 0 0 0 0 WQ150 0 0 27 0 1 58 2 0 WQ180 0 0 4 0 0 3 11 1 WQ210 0 9 8 0 3 13 7 0 WQ235 0 0 2 0 0 2 3 0 WQ245 0 0 38 0 5 3 2 0 WQ250 2 0 15 0 2 9 1 0 WQ260 0 0 18 1 0 1 2 0 WQ265 0 0 33 0 0 2 0 0 WQ310 0 0 5 0 0 4 1 0 WQ320 0 0 10 0 0 0 4 0 WQ370 0 0 1 0 0 1 2 0 WQ380 0 0 0 0 0 0 4 0 WQ400 0 0 1 0 0 0 2 2 WQ420 0 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 0 WQ445 0 0 2 0 0 9 0 0 WQ455 0 0 0 0 0 0 2 0 WQ465 0 0 0 0 0 0 1 0 WQ480 0 0 1 0 0 1 2 0 MC230 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 MC350 0 0 0 0 0 1 0 0 MC360 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 0 MC410 0 0 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 RQ8.0 0 0 7 0 0 1 0 0 RQ8.2 0 0 10 0 0 3 0 0 RQ8.3 0 0 5 0 0 0 0 0 RQ8.4 0 0 4 0 0 329 0 0 218

RQ8.5 0 0 1 0 0 94 0 0 RQ8.6 0 0 1 2 0 73 0 0 RQ8.8 0 0 1 2 0 17 0 4 RQ8.9 0 0 4 0 0 3 0 1 RQ9.0 0 0 1 0 0 95 0 0 RQ9.1 0 0 0 1 0 6 0 0 RQ9.2 0 0 0 1 0 2 0 0 RQ9.3 0 0 0 0 0 137 0 0 RQ9.4 0 0 0 0 0 60 1 1 RQ9.5 0 0 0 0 0 2 0 0 RQ9.6 0 0 4 0 0 0 1 2 TB 0 5 2 0 0 162 5 3 Dictyophycus Dinomischus Diraphora Echmatocrinus Ehmaniella strata Dalyia sp. Diagoniella hindei Echinodermata gracilis isolatus bellicostata brachiatus spp. WQ+120 0 0 0 0 2 0 0 21 WQ0 0 0 3 0 3 0 0 11 WQ110 2 0 1 0 10 1 1 3 WQ120 8 3 112 0 12 2 1 4 WQ130 3 0 16 0 44 0 0 2 WQ150 4 1 1 3 54 2 0 35 WQ180 4 2 8 0 9 0 0 10 WQ210 1 8 2 7 139 6 0 18 WQ235 8 2 94 0 38 0 0 4 WQ245 2 61 20 0 54 1 0 14 WQ250 1 39 6 0 18 0 0 7 WQ260 0 4 5 0 76 0 0 19 WQ265 2 3 27 0 25 0 0 9 WQ310 0 3 1 0 21 0 0 16 WQ320 0 0 0 1 10 0 0 12 WQ370 1 0 0 0 14 0 0 26 WQ380 0 0 0 0 1 0 0 4 WQ400 0 1 12 1 62 1 0 23 WQ420 0 0 0 1 14 0 0 3 WQ430 0 0 0 0 6 0 0 2 WQ445 0 0 0 0 13 0 3 41 WQ455 0 0 0 0 2 0 0 5 WQ465 0 0 0 0 9 0 0 5 WQ480 1 0 0 0 1 0 0 13 MC230 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 0 MC410 0 0 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 219

RQ8.0 0 0 0 0 7 0 0 0 RQ8.2 0 0 0 0 9 0 7 1 RQ8.3 0 0 0 0 8 0 1 0 RQ8.4 0 0 0 0 15 0 2 1 RQ8.5 0 0 0 0 7 0 0 0 RQ8.6 0 0 0 0 3 0 0 0 RQ8.8 0 0 0 0 1 0 0 0 RQ8.9 0 0 0 0 1 0 0 0 RQ9.0 0 0 0 0 9 0 0 1 RQ9.1 0 0 0 0 1 0 1 0 RQ9.2 0 0 0 0 1 0 0 0 RQ9.3 0 0 0 0 1 0 0 0 RQ9.4 0 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 0 0 0 RQ9.6 0 0 0 0 0 0 0 0 TB 0 15 0 6 17 0 0 5 Elrathia Elrathina Fieldospongia strata Eiffelia globosa Eldonia ludwigi Emeraldella brocki Falospongia falata Fieldia lanceolata permulta cordillerae bellilineata WQ+120 0 89 0 4 0 0 0 0 WQ0 0 25 0 0 8 0 1 0 WQ110 1 0 0 0 1 0 0 0 WQ120 1 0 0 0 3 2 9 0 WQ130 1 1 0 0 1 0 0 0 WQ150 6 0 0 3 0 1 0 0 WQ180 3 0 0 0 0 1 1 1 WQ210 10 0 2 0 1 1 0 3 WQ235 2 0 0 13 2 0 0 0 WQ245 17 4 0 461 0 0 0 0 WQ250 4 0 2 324 0 0 0 0 WQ260 0 5 0 77 0 0 0 0 WQ265 2 1 0 10 1 0 0 0 WQ310 0 0 1 0 0 0 0 0 WQ320 2 0 1 0 1 0 0 0 WQ370 0 1 3 6 0 0 0 0 WQ380 0 0 0 49 0 0 0 0 WQ400 2 0 1 66 2 0 5 0 WQ420 0 1 0 329 0 0 0 0 WQ430 0 0 0 85 0 0 0 0 WQ445 0 1 0 53 0 0 0 0 WQ455 0 0 0 3 1 0 0 0 WQ465 0 1 0 387 0 0 0 0 WQ480 0 0 0 4 0 0 0 0 MC230 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 MC380 0 0 0 0 2 0 0 0 MC390 0 0 0 0 0 0 0 0 MC400 0 0 0 0 1 0 0 0 MC410 0 0 0 0 1 0 0 0 220

MC420 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 RQ8.0 0 0 0 0 0 0 0 0 RQ8.2 0 0 0 0 0 0 0 0 RQ8.3 0 0 0 0 7 0 0 0 RQ8.4 0 0 0 0 2 0 0 0 RQ8.5 0 0 0 0 1 0 0 0 RQ8.6 0 0 0 0 1 0 0 0 RQ8.8 0 0 0 0 1 0 0 0 RQ8.9 0 0 0 0 0 0 0 0 RQ9.0 0 0 0 0 1 1 0 0 RQ9.1 0 0 0 0 1 1 0 0 RQ9.2 0 0 0 0 1 0 0 0 RQ9.3 0 0 0 0 0 1 0 0 RQ9.4 0 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 1 0 0 0 RQ9.6 0 0 0 0 1 0 0 0 TB 12 0 0 33 0 1 0 0 Habelia? Halicondrites Hamptonia Haplophrentis strata Gogia? sp. Habelia optata Hallucigenia spp. Hanburia gloriosa brevicauda elissa bowerbanki carinatus WQ+120 0 3 0 0 6 1 0 8 WQ0 0 0 0 0 0 0 0 2 WQ110 0 1 0 0 14 0 0 4 WQ120 0 1 0 0 7 2 0 8 WQ130 0 0 0 0 15 0 0 4 WQ150 0 3 1 0 5 0 0 27 WQ180 0 0 0 0 6 1 1 8 WQ210 2 3 0 2 46 0 1 29 WQ235 0 0 2 2 2 0 1 12 WQ245 0 0 0 2 0 41 1 13 WQ250 2 0 0 0 0 1 0 8 WQ260 0 2 19 0 4 0 0 10 WQ265 0 0 6 0 1 0 0 2 WQ310 0 2 1 0 2 0 0 3 WQ320 0 0 0 0 0 0 0 5 WQ370 0 0 0 1 0 0 0 3 WQ380 0 1 1 0 0 0 0 0 WQ400 0 5 1 0 0 0 0 7 WQ420 0 0 0 0 0 0 0 8 WQ430 0 0 0 0 0 0 0 1 WQ445 0 0 0 0 0 2 0 1 WQ455 0 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 7 WQ480 0 0 1 0 0 0 0 4 MC230 0 0 0 0 0 0 0 182 MC240 0 0 0 0 0 0 0 329 MC300 0 0 0 0 0 0 0 185 MC310 0 0 0 0 0 0 0 138 MC320 0 0 0 0 0 0 0 173 MC330 0 0 0 0 0 0 0 139 MC340 0 0 0 0 0 0 0 120 MC350 0 0 0 0 0 0 0 302 MC360 0 0 0 0 0 0 0 113 MC370 0 0 0 0 0 0 0 365 221

MC380 0 0 0 0 0 0 0 396 MC390 0 0 0 0 1 0 0 65 MC400 0 0 0 0 0 0 0 309 MC410 0 0 0 0 0 0 0 326 MC420 0 0 0 0 0 0 0 19 MC480 0 0 0 0 0 0 0 28 MC490 0 0 0 0 0 0 0 15 MC500 0 0 0 0 0 0 0 1 RQ8.0 0 0 0 0 0 0 0 15 RQ8.2 0 0 0 0 0 1 0 34 RQ8.3 0 0 0 0 0 0 0 17 RQ8.4 0 0 0 0 0 0 0 29 RQ8.5 0 2 0 0 0 1 0 10 RQ8.6 0 0 0 0 0 1 0 16 RQ8.8 0 0 0 0 0 2 0 7 RQ8.9 0 0 0 0 0 2 0 15 RQ9.0 0 0 0 0 0 0 0 19 RQ9.1 0 1 0 0 0 2 0 6 RQ9.2 0 1 0 0 0 2 0 8 RQ9.3 1 0 0 0 0 0 0 9 RQ9.4 0 0 0 0 0 0 0 3 RQ9.5 0 0 0 0 0 1 0 8 RQ9.6 0 0 0 0 0 1 0 9 TB 0 0 37 8 119 90 5 0 Hazelia Helmetia Herpetogaster Hurdia Insolicorypha Isoxys Laggania Leanchoilia strata Kootenia burgessensis spp. expansa collinsi sp. psygma? spp. cambria spp. WQ+120 9 0 0 3 0 2 0 1 1 WQ0 10 0 0 1 0 3 0 1 1 WQ110 27 0 0 7 0 1 4 0 0 WQ120 133 0 0 2 1 1 1 10 3 WQ130 23 0 0 0 0 1 1 2 5 WQ150 140 0 1 4 0 21 5 19 1 WQ180 303 0 0 5 0 12 6 8 1 WQ210 197 5 1 12 0 9 7 28 1 WQ235 412 0 0 1 0 12 18 4 2 WQ245 228 0 0 1 0 58 3 5 0 WQ250 250 0 1 1 0 5 15 2 0 WQ260 351 0 1 2 0 1 3 5 23 WQ265 306 0 1 5 0 2 3 3 2 WQ310 112 0 0 2 0 3 2 4 0 WQ320 90 0 0 5 0 2 0 2 0 WQ370 171 0 0 3 0 6 8 1 1 WQ380 43 0 0 0 0 0 0 2 0 WQ400 302 0 0 6 0 5 7 2 2 WQ420 142 0 0 0 0 1 10 0 0 WQ430 46 0 0 1 0 1 5 2 0 WQ445 1028 0 0 0 0 0 1 0 0 WQ455 12 0 0 0 0 0 0 0 0 WQ465 191 0 0 0 0 0 3 1 0 WQ480 28 0 1 2 0 0 0 1 11 MC230 0 0 0 0 0 1 0 0 0 MC240 0 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 0 MC330 0 0 0 1 0 1 0 0 0 222

MC340 0 0 0 2 0 2 1 0 0 MC350 0 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 2 0 0 0 MC370 0 0 0 1 0 0 0 0 0 MC380 0 0 0 0 0 2 0 0 0 MC390 0 0 0 0 0 1 3 0 0 MC400 0 0 0 2 0 0 4 0 0 MC410 1 0 0 1 0 0 8 0 0 MC420 0 0 0 0 0 0 3 0 0 MC480 0 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 1 0 0 0 MC500 0 0 0 0 0 0 0 0 0 RQ8.0 0 0 1 12 0 5 0 0 37 RQ8.2 1 0 6 7 0 9 1 0 77 RQ8.3 0 3 4 7 0 2 0 0 46 RQ8.4 7 0 5 9 0 11 1 0 61 RQ8.5 2 0 12 10 0 12 0 0 46 RQ8.6 0 3 8 9 0 1 1 0 30 RQ8.8 2 3 8 19 0 4 0 0 68 RQ8.9 0 1 8 13 0 5 1 0 41 RQ9.0 0 0 7 14 0 3 0 0 50 RQ9.1 1 0 3 6 0 2 0 0 21 RQ9.2 0 0 4 13 0 7 0 0 36 RQ9.3 0 0 4 21 0 9 0 0 70 RQ9.4 0 0 3 10 0 10 0 0 41 RQ9.5 0 0 3 12 0 5 0 0 38 RQ9.6 0 1 1 7 0 4 0 0 65 TB 65 1 0 41 0 14 13 22 53 Leptomitus Liangshanella Lingulella Mackenzia Marrella strata Louisella pedunculata Marpolia spissa lineatus spp. waptaensis costalis splendens WQ+120 0 24 10 0 0 2 119 WQ0 0 53 19 8 0 2 571 WQ110 0 134 2 1 0 1 14 WQ120 3 167 8 13 0 0 333 WQ130 1 107 2 17 0 0 1425 WQ150 67 328 46 0 2 0 172 WQ180 4 49 5 0 0 0 40 WQ210 9 295 6 2 3 2 237 WQ235 1 287 5 1 1 4 156 WQ245 2 1400 172 0 0 0 496 WQ250 1 833 9 0 0 1 80 WQ260 0 1090 0 1 0 0 171 WQ265 0 420 1 0 1 0 41 WQ310 21 52 7 0 1 21 22 WQ320 5 13 5 1 0 3 186 WQ370 20 19 2 0 0 0 4 WQ380 0 70 2 0 1 0 13 WQ400 2 68 0 3 0 0 265 WQ420 0 475 0 0 0 0 247 WQ430 0 40 4 0 0 0 36 WQ445 0 98 2 0 1 0 1 WQ455 0 13 0 0 0 0 119 WQ465 0 13 11 0 0 0 0 WQ480 0 25 3 0 1 4 121 MC230 0 25 4 0 0 0 0 MC240 0 37 1 0 0 0 0 223

MC300 0 44 0 0 0 0 0 MC310 0 18 0 0 0 0 0 MC320 0 25 0 0 0 0 0 MC330 0 71 0 0 0 0 0 MC340 0 63 0 0 0 0 0 MC350 0 306 0 0 0 0 0 MC360 0 197 0 0 0 0 0 MC370 0 476 0 0 0 0 0 MC380 0 851 0 0 0 0 0 MC390 0 355 0 0 0 0 0 MC400 0 897 0 0 0 0 0 MC410 0 1318 0 0 0 0 0 MC420 0 446 0 0 0 0 0 MC480 0 11 0 0 0 0 0 MC490 0 2 0 0 0 0 0 MC500 0 6 0 0 0 0 0 RQ8.0 0 40 1 1 1 0 0 RQ8.2 0 61 0 0 1 0 0 RQ8.3 0 51 0 0 0 0 0 RQ8.4 0 40 2 0 1 0 0 RQ8.5 1 45 0 0 0 0 0 RQ8.6 0 7 1 0 2 0 0 RQ8.8 0 75 0 0 3 0 0 RQ8.9 0 40 0 0 4 0 0 RQ9.0 0 56 0 0 2 0 0 RQ9.1 0 74 0 0 2 0 0 RQ9.2 0 96 0 0 6 0 0 RQ9.3 0 73 0 0 4 0 0 RQ9.4 0 47 0 0 1 1 0 RQ9.5 0 65 0 0 1 0 0 RQ9.6 0 187 0 0 5 0 0 TB 56 0 9 0 11 693 23 Micromitra Molaria Mollisonia Nisusia Odontogriphus strata Morania sp. Naraoia spp. Odaraia alata burgessensis spinifera sp. burgessensis omalus WQ+120 0 0 0 48 0 0 0 1 WQ0 0 8 0 224 3 0 1 0 WQ110 1 3 0 5 1 4 0 0 WQ120 16 28 0 337 30 4 7 4 WQ130 6 19 0 183 9 3 9 0 WQ150 44 20 16 115 9 16 3 0 WQ180 5 0 1 103 17 6 2 0 WQ210 12 2 2 271 65 32 11 0 WQ235 5 6 1 99 115 5 23 5 WQ245 13 0 0 105 1 15 0 0 WQ250 8 1 0 26 3 10 2 0 WQ260 19 13 0 56 4 12 93 125 WQ265 3 5 0 14 8 2 10 12 WQ310 1 0 0 7 8 1 7 0 WQ320 1 10 0 26 39 4 9 6 WQ370 3 1 0 19 0 5 0 0 WQ380 0 1 0 66 1 0 1 2 WQ400 1 4 0 772 49 3 8 6 WQ420 1 4 0 6 0 2 1 5 WQ430 0 0 1 0 1 0 0 0 WQ445 2 0 0 6 1 0 0 0 WQ455 1 6 0 2 10 0 0 0 224

WQ465 6 0 0 0 0 1 0 0 WQ480 6 0 0 1 11 2 23 2 MC230 0 0 2 0 0 0 0 0 MC240 0 0 2 0 0 0 1 0 MC300 0 1 0 0 0 0 0 0 MC310 1 0 0 0 0 0 0 0 MC320 0 0 2 0 0 0 0 0 MC330 0 1 0 0 0 0 0 0 MC340 0 3 1 0 0 0 0 0 MC350 0 3 3 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 MC370 0 4 3 0 0 0 0 0 MC380 0 10 8 0 0 0 0 0 MC390 0 3 0 0 0 0 0 2 MC400 0 2 3 0 0 0 0 0 MC410 0 4 3 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 MC480 0 1 3 0 0 0 0 0 MC490 0 0 1 0 0 0 0 0 MC500 0 0 2 0 0 0 0 0 RQ8.0 27 0 1 0 0 6 2 0 RQ8.2 74 0 0 0 0 4 1 0 RQ8.3 29 0 0 0 0 2 1 0 RQ8.4 111 0 0 0 2 1 1 0 RQ8.5 39 0 0 0 0 1 0 0 RQ8.6 12 0 0 0 0 1 0 0 RQ8.8 27 0 0 0 1 2 0 0 RQ8.9 50 0 0 0 0 1 0 0 RQ9.0 42 0 0 0 1 0 0 0 RQ9.1 38 1 0 0 1 0 1 0 RQ9.2 39 1 0 0 0 0 1 0 RQ9.3 24 0 0 0 0 0 1 0 RQ9.4 60 0 0 0 1 0 0 0 RQ9.5 13 0 0 0 0 0 0 0 RQ9.6 48 1 0 0 0 2 0 0 TB 16 4 1 0 35 23 1 1 Opabinia Orthrozanclus Oryctocephalus Ottoia Paterina Peronochaeta strata Olenoides serratus Pagetia bootes regalis reburrus spp. prolifica zenobia dubia WQ+120 0 0 0 0 11 4 1 0 WQ0 0 2 0 0 6 0 1 16 WQ110 0 0 0 0 6 194 0 0 WQ120 4 0 0 3 272 164 0 1 WQ130 1 0 1 1 27 7 7 1 WQ150 14 0 3 34 3 106 5 0 WQ180 12 0 2 11 10 4 0 0 WQ210 20 0 1 32 35 65 0 0 WQ235 20 0 0 5 63 33 1 0 WQ245 37 0 0 4 3 248 1 0 WQ250 8 0 1 12 12 98 1 0 WQ260 10 0 0 6 46 51 1 0 WQ265 9 0 0 5 22 11 0 0 WQ310 1 0 0 84 4 1 0 0 WQ320 0 0 0 9 16 1 0 0 WQ370 7 0 0 1 0 21 0 0 WQ380 3 0 0 0 9 2 0 0 WQ400 10 0 1 8 56 2 0 1 225

WQ420 21 1 0 0 12 0 0 0 WQ430 3 0 0 1 1 2 0 0 WQ445 7 0 0 0 0 1 0 0 WQ455 0 0 0 0 1 0 0 0 WQ465 8 0 0 0 0 2 0 0 WQ480 2 0 0 0 8 0 0 0 MC230 0 1 0 0 0 0 0 0 MC240 0 5 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 MC330 0 1 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 MC350 0 1 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 MC380 0 1 0 0 0 0 0 0 MC390 0 2 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 0 MC410 0 2 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 1 0 MC490 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 RQ8.0 4 0 0 0 21 7 1 0 RQ8.2 10 2 0 6 31 4 0 0 RQ8.3 3 1 0 0 79 2 0 0 RQ8.4 6 0 0 2 57 7 0 0 RQ8.5 0 1 0 2 27 3 0 0 RQ8.6 1 1 0 0 31 1 0 0 RQ8.8 1 1 0 1 74 1 0 0 RQ8.9 3 0 0 3 43 8 0 0 RQ9.0 0 2 0 2 86 2 0 0 RQ9.1 1 1 0 1 58 5 0 0 RQ9.2 1 1 0 0 70 3 0 0 RQ9.3 3 0 0 0 125 1 0 0 RQ9.4 2 0 0 0 51 1 0 0 RQ9.5 2 0 0 0 56 1 0 0 RQ9.6 9 0 0 0 169 0 0 0 TB 0 0 0 75 68 0 33 0 Petaloptyon Pirania Plenocaris Pollingeria Priscansermarinus strata Perspicaris spp. Pikaia gracilens Portalia mira sp. muricata plena grandis barnetti WQ+120 8 0 0 0 0 0 0 0 WQ0 5 0 3 0 25 0 0 0 WQ110 10 0 0 0 0 10 0 1 WQ120 23 0 4 5 24 4 3 0 WQ130 5 0 6 1 4 0 0 0 WQ150 11 0 0 49 3 0 3 3 WQ180 8 0 0 12 5 16 0 0 WQ210 7 0 1 44 1 133 3 0 WQ235 26 0 1 4 4 3 2 0 WQ245 0 0 0 11 1 3 0 0 WQ250 0 0 0 10 0 29 1 0 WQ260 31 0 0 9 10 7 1 0 WQ265 13 0 0 4 7 376 1 0 WQ310 1 0 0 5 0 47 1 0 226

WQ320 3 1 1 5 5 5 0 0 WQ370 0 0 0 9 0 3 0 0 WQ380 5 0 0 1 1 51 0 0 WQ400 21 0 0 5 8 58 0 0 WQ420 0 0 0 0 0 5 1 0 WQ430 0 0 0 0 0 85 0 0 WQ445 6 0 0 6 0 122 0 0 WQ455 7 0 0 0 0 1 0 0 WQ465 0 0 0 2 0 0 0 0 WQ480 2 0 0 1 1 1 0 0 MC230 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 MC330 3 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 MC350 2 0 0 0 0 0 0 0 MC360 1 0 0 0 0 0 0 0 MC370 0 0 0 0 1 0 0 0 MC380 3 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 0 MC410 0 0 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 RQ8.0 0 0 0 0 0 12 0 0 RQ8.2 0 0 0 0 0 66 0 0 RQ8.3 0 0 0 0 0 20 0 0 RQ8.4 0 0 0 0 0 52 0 0 RQ8.5 0 0 0 0 0 81 0 0 RQ8.6 0 0 0 0 0 20 0 0 RQ8.8 0 0 0 1 0 43 0 1 RQ8.9 0 0 0 0 0 32 0 0 RQ9.0 0 0 0 0 0 115 0 0 RQ9.1 0 0 0 0 0 170 0 14 RQ9.2 0 0 0 0 0 126 0 0 RQ9.3 0 0 0 0 0 132 0 16 RQ9.4 0 0 0 0 0 75 1 51 RQ9.5 0 0 0 0 0 30 0 1 RQ9.6 0 0 0 0 0 88 0 0 TB 40 8 0 27 1 0 0 0 Ptychagnostus Sarotrocercus Selkirkia Sidneyia Spartobranchus strata Protospongia hicksi Scenella amii Skania fragilis praecurrens oblita spp. inexpectans tenuis WQ+120 0 10 0 13 39 1 0 8 WQ0 0 24 2 52 29 0 0 20 WQ110 2 38 0 11 4 0 0 2 WQ120 8 106 3 56 40 16 0 248 WQ130 0 156 1 249 10 5 0 39 WQ150 15 757 7 164 34 9 0 0 WQ180 2 393 5 16 50 11 0 1 WQ210 11 1011 4 99 191 54 0 12 WQ235 0 175 0 24 104 6 0 2 WQ245 37 238 0 82 301 1 0 0 227

WQ250 15 91 0 32 316 2 0 1 WQ260 0 203 1 71 122 2 1 752 WQ265 0 140 2 43 56 4 0 111 WQ310 7 115 0 26 18 1 0 5 WQ320 0 64 1 12 29 2 0 2 WQ370 0 76 1 24 12 0 0 1 WQ380 0 20 0 0 20 5 0 11 WQ400 2 41 1 57 79 12 0 24 WQ420 0 10 0 62 22 1 0 3 WQ430 0 15 0 18 4 1 0 2 WQ445 1 72 0 4 9 1 0 0 WQ455 0 0 0 7 1 1 0 2 WQ465 0 19 0 10 2 0 0 0 WQ480 0 8 0 10 4 1 0 1 MC230 0 0 0 0 0 1 0 0 MC240 0 0 0 0 0 2 0 0 MC300 0 0 0 0 0 4 0 0 MC310 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 5 0 0 MC330 0 0 0 0 0 6 0 0 MC340 0 0 0 0 1 3 0 0 MC350 0 0 0 0 1 12 0 0 MC360 0 0 0 0 2 8 0 0 MC370 0 0 0 0 1 9 0 0 MC380 0 0 0 0 1 71 0 0 MC390 0 0 0 0 0 11 0 0 MC400 0 0 0 0 1 36 0 1 MC410 0 2 0 0 0 31 0 0 MC420 0 0 0 0 0 13 0 0 MC480 0 1 0 0 0 2 0 9 MC490 0 23 0 0 0 3 0 13 MC500 0 26 0 0 0 3 0 16 RQ8.0 0 7 0 0 0 13 0 0 RQ8.2 0 1 0 0 1 12 2 0 RQ8.3 0 0 0 0 0 6 2 0 RQ8.4 0 9 0 0 1 25 2 0 RQ8.5 0 5 0 0 0 14 2 0 RQ8.6 0 4 0 0 0 24 5 0 RQ8.8 0 5 0 0 0 26 0 0 RQ8.9 0 5 0 0 0 19 0 1 RQ9.0 0 11 0 1 0 30 0 0 RQ9.1 0 1 0 1 0 13 2 0 RQ9.2 0 0 0 1 0 18 0 0 RQ9.3 0 0 0 0 0 21 2 0 RQ9.4 0 0 0 0 0 16 0 0 RQ9.5 0 0 0 0 0 8 0 0 RQ9.6 0 1 0 0 0 12 0 0 TB 13 0 0 7 10 19 0 0 Stephenoscolex Takakkawia Tubullela Tuzoia Undet Algae Undet Arthropoda Undet Arthropoda strata argutus lineata sp. spp. WQ-A WQ-D WQ-A WQ+120 0 0 0 0 2 0 0 WQ0 0 1 0 0 1 0 0 WQ110 0 1 6 0 0 0 0 WQ120 0 21 4 0 1 1 1 WQ130 0 11 3 2 2 1 1 WQ150 0 12 49 1 6 0 0 228

WQ180 0 0 4 1 4 0 0 WQ210 0 634 4 3 25 0 0 WQ235 0 77 3 0 7 0 0 WQ245 0 6 5 2 58 0 0 WQ250 0 5 5 1 20 0 0 WQ260 0 39 0 2 1 0 1 WQ265 0 3 1 0 1 0 0 WQ310 0 7 2 1 3 0 0 WQ320 0 409 4 0 1 0 0 WQ370 0 2 5 1 2 0 0 WQ380 2 2 1 0 0 0 0 WQ400 2 37 2 0 2 0 0 WQ420 10 0 2 0 0 0 0 WQ430 1 0 0 0 0 0 0 WQ445 1 17 3 0 1 0 0 WQ455 128 0 0 0 0 0 0 WQ465 0 0 4 0 0 0 0 WQ480 3 2 0 0 6 0 0 MC230 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 MC330 0 0 0 2 0 0 0 MC340 0 0 0 2 0 0 0 MC350 0 0 0 2 0 0 0 MC360 0 0 0 3 0 0 0 MC370 0 0 0 0 0 0 0 MC380 0 0 0 6 0 0 0 MC390 0 0 0 4 0 0 0 MC400 0 0 0 2 0 0 0 MC410 0 0 0 9 0 0 0 MC420 0 0 0 1 0 0 0 MC480 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 RQ8.0 0 0 24 3 0 0 0 RQ8.2 0 5 33 4 0 0 0 RQ8.3 0 1 20 6 0 0 0 RQ8.4 0 0 63 2 0 0 0 RQ8.5 0 0 18 2 0 0 0 RQ8.6 0 0 10 4 0 0 0 RQ8.8 0 1 12 7 0 0 0 RQ8.9 0 0 20 1 0 0 0 RQ9.0 0 0 21 1 0 0 0 RQ9.1 0 8 8 3 0 0 0 RQ9.2 0 17 14 3 0 0 0 RQ9.3 0 5 6 3 0 0 0 RQ9.4 0 6 8 2 0 0 0 RQ9.5 0 2 7 4 0 0 0 RQ9.6 0 0 11 3 0 0 0 TB 0 8 299 7 0 0 0 Undet Arthropoda Undet Arthropoda Undet Brachiopoda Undet Chaetognatha Undet Holothuroidea Undet Polychaeta WQ- Undet Porifera Undet Porifera strata WQ-B WQ-C WQ-A WQ-A WQ-A B WQ-A WQ-B WQ+120 0 0 0 0 0 0 0 0 WQ0 1 0 0 0 0 0 0 0 229

WQ110 0 0 0 0 0 1 0 0 WQ120 1 0 0 4 1 5 0 0 WQ130 0 0 0 0 0 1 0 0 WQ150 0 0 3 1 0 0 20 0 WQ180 1 0 0 0 0 0 125 0 WQ210 0 2 0 3 0 0 0 1 WQ235 2 0 0 4 0 0 2 0 WQ245 0 0 0 0 0 0 0 0 WQ250 0 0 0 1 0 0 0 0 WQ260 0 0 0 4 0 80 0 0 WQ265 0 0 0 2 1 0 0 0 WQ310 0 0 2 0 0 0 0 0 WQ320 0 0 0 2 0 0 0 0 WQ370 0 0 0 0 0 2 0 0 WQ380 0 0 0 0 0 0 0 0 WQ400 0 0 0 2 0 4 0 2 WQ420 0 0 0 2 0 1 0 0 WQ430 0 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 0 WQ480 0 0 0 1 0 0 0 0 MC230 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 0 MC410 0 0 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 RQ8.0 0 0 0 0 0 0 0 0 RQ8.2 0 0 0 0 0 0 0 0 RQ8.3 0 0 0 0 0 0 0 0 RQ8.4 0 0 0 0 0 0 0 0 RQ8.5 0 0 0 0 0 0 0 0 RQ8.6 0 0 0 0 0 0 0 0 RQ8.8 0 0 0 0 0 0 0 0 RQ8.9 0 0 0 0 0 0 0 0 RQ9.0 0 0 0 0 0 0 0 0 RQ9.1 0 0 0 0 0 0 0 0 RQ9.2 0 0 0 0 0 0 0 0 RQ9.3 0 0 0 0 0 0 0 0 RQ9.4 0 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 0 0 0 RQ9.6 0 0 0 0 0 0 0 0 TB 0 0 0 0 0 0 0 0 230 strata Undet Taxon A Undet Taxon WQ-A Undet Taxon WQ-B Undet Taxon WQ-C Undet Taxon WQ-D Undet Taxon WQ-E Undet Taxon WQ-F WQ+120 0 0 0 0 0 0 0 WQ0 0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 0 WQ120 0 0 2 1 1 0 0 WQ130 1 1 0 0 0 0 0 WQ150 0 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 0 WQ210 31 0 0 1 0 0 5 WQ235 0 1 0 0 0 0 0 WQ245 1 0 0 0 0 0 0 WQ250 41 0 0 0 0 0 3 WQ260 0 0 0 0 0 2 0 WQ265 10 0 0 0 0 0 0 WQ310 186 0 0 0 0 0 0 WQ320 1 0 0 0 1 0 0 WQ370 19 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 WQ400 4 0 0 1 4 0 0 WQ420 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 WQ480 0 0 0 0 2 0 0 MC230 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 MC410 0 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 RQ8.0 0 0 0 0 0 0 0 RQ8.2 0 0 0 0 0 0 0 RQ8.3 0 0 0 0 0 0 0 RQ8.4 0 0 0 0 0 0 0 RQ8.5 0 0 0 0 0 0 0 RQ8.6 0 0 0 0 0 0 0 RQ8.8 0 0 0 0 0 0 0 RQ8.9 0 0 0 0 0 0 0 RQ9.0 0 0 0 0 0 0 0 RQ9.1 0 0 0 0 0 0 0 RQ9.2 0 0 0 0 0 0 0 RQ9.3 0 0 0 0 0 0 0 231

RQ9.4 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 0 0 RQ9.6 0 0 0 0 0 0 0 TB 0 0 0 0 0 0 0 Undet Taxon Undet Taxon Undet Taxon Vauxia Wahpia Walcottidiscus Waptia Waputikia strata Wapkia grandis WQ-H WQ-I WQ-J sp. sp. sp. fieldensis ramosa WQ+120 0 0 0 0 1 0 0 3 1 WQ0 0 0 0 0 0 0 0 28 0 WQ110 0 0 0 1 0 0 2 1 0 WQ120 0 2 2 0 3 0 0 44 0 WQ130 0 1 0 1 0 0 0 101 0 WQ150 0 0 1 1 0 1 5 41 1 WQ180 0 0 0 0 0 0 2 2 2 WQ210 0 0 1 14 4 8 1 97 5 WQ235 0 0 34 0 2 0 1 19 0 WQ245 0 0 4 1 3 0 11 5 0 WQ250 0 0 13 0 9 1 2 8 1 WQ260 0 0 1 0 0 1 1 40 0 WQ265 0 0 1 0 0 1 0 9 0 WQ310 0 0 0 0 0 0 0 3 0 WQ320 0 0 0 1 0 0 0 17 0 WQ370 0 0 0 0 0 0 1 0 0 WQ380 0 0 0 0 0 0 0 4 0 WQ400 2 0 4 6 3 0 1 3 0 WQ420 0 0 0 0 1 0 0 6 0 WQ430 0 0 0 0 0 0 0 1 0 WQ445 0 0 0 0 0 2 1 1 0 WQ455 0 0 0 0 0 0 0 1 0 WQ465 0 0 0 0 1 0 0 0 0 WQ480 0 0 0 0 4 1 0 2 0 MC230 0 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 0 0 MC410 0 0 0 1 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 0 RQ8.0 0 0 0 40 0 0 0 1 0 RQ8.2 0 0 0 110 0 0 0 9 0 RQ8.3 0 0 0 61 0 0 0 2 0 RQ8.4 0 0 0 180 0 0 0 9 0 RQ8.5 0 0 0 115 0 0 0 6 0 RQ8.6 0 0 0 70 0 0 0 3 0 RQ8.8 0 0 0 99 0 0 1 3 0 RQ8.9 0 0 0 70 0 0 0 1 0 232

RQ9.0 0 0 0 147 0 0 0 5 0 RQ9.1 0 0 0 60 0 0 0 3 0 RQ9.2 0 0 0 76 0 0 0 3 0 RQ9.3 0 0 0 98 0 0 0 1 0 RQ9.4 0 0 0 82 0 0 0 2 0 RQ9.5 0 0 0 49 0 0 0 3 0 RQ9.6 0 0 0 78 0 0 0 1 0 TB 0 0 0 57 96 0 196 2 0 Wiwaxia Yohoia Yuknessia New strata Oesia Acrotretid Metaspriggina Peronopsis Hurdia Fuxianospira Cyanobacteria corrugata tenuis simplex dinocariidid B WQ+120 1 4 0 0 0 0 0 0 0 0 0 WQ0 1 46 2 0 0 0 0 0 0 0 0 WQ110 1 19 1 0 0 0 0 0 0 0 0 WQ120 1 37 4 2 0 0 0 0 0 0 0 WQ130 1 47 1 2 0 1 0 0 0 0 0 WQ150 1 7 0 0 0 0 0 0 0 0 0 WQ180 28 11 4 0 0 0 0 0 0 0 0 WQ210 31 55 0 0 0 0 0 0 0 0 0 WQ235 51 84 4 0 0 0 0 0 0 0 0 WQ245 44 4 3 0 0 0 0 0 0 0 0 WQ250 7 20 1 0 0 0 0 0 0 0 0 WQ260 25 26 1 0 0 0 0 0 0 0 0 WQ265 2 13 2 0 0 0 0 0 0 0 0 WQ310 15 7 0 0 0 0 0 0 0 0 0 WQ320 14 106 0 0 0 1 0 0 0 0 0 WQ370 3 3 0 0 0 0 0 0 0 0 0 WQ380 5 0 0 0 0 0 0 0 0 0 0 WQ400 29 38 0 0 0 4 0 0 0 0 0 WQ420 2 138 0 0 0 0 0 0 0 0 0 WQ430 7 23 0 0 0 0 0 0 0 0 0 WQ445 5 1 0 0 0 0 0 0 0 0 0 WQ455 31 6 0 0 0 0 0 0 0 0 0 WQ465 1 0 0 0 0 0 0 0 0 0 0 WQ480 1 1 0 0 0 2 0 0 0 0 0 MC230 0 0 0 69 3 2 27 0 0 3 0 MC240 0 0 0 81 5 6 41 0 3 7 0 MC300 0 0 0 248 0 0 11 0 0 2 1 MC310 0 0 0 93 8 0 28 0 0 2 0 MC320 0 0 0 94 6 0 32 0 0 1 0 MC330 0 0 0 132 0 4 100 1 0 0 0 MC340 0 0 0 126 9 1 51 2 0 1 0 MC350 0 0 0 262 4 5 73 0 0 3 3 MC360 0 0 0 84 0 11 62 0 0 0 0 MC370 0 0 0 294 0 28 94 1 0 6 0 MC380 0 0 0 307 0 79 256 0 1 2 1 MC390 0 0 0 200 0 9 94 0 2 1 0 MC400 0 0 0 507 0 11 237 2 0 0 3 MC410 0 0 0 92 0 25 180 1 2 1 2 MC420 0 0 0 51 0 4 47 0 2 18 0 MC480 0 0 0 10 0 18 243 0 0 0 0 MC490 0 0 0 7 0 10 370 0 2 0 0 MC500 0 0 0 3 0 8 441 0 2 0 0 RQ8.0 0 0 0 0 2 0 0 0 0 0 0 RQ8.2 0 0 0 0 7 0 0 0 4 0 0 RQ8.3 0 0 0 0 0 0 1 0 5 0 0 RQ8.4 0 0 0 0 3 0 3 0 22 0 0 233

RQ8.5 0 0 0 0 0 0 0 0 26 0 0 RQ8.6 0 0 0 0 1 0 0 0 6 0 0 RQ8.8 0 0 0 0 0 0 0 0 12 0 0 RQ8.9 0 0 0 0 0 0 0 0 15 0 0 RQ9.0 0 0 0 1 0 0 0 0 12 0 0 RQ9.1 0 0 0 0 0 0 0 0 15 0 0 RQ9.2 0 0 0 0 0 0 0 0 23 0 0 RQ9.3 0 0 0 0 0 0 0 0 15 0 0 RQ9.4 0 0 0 0 0 0 0 0 7 0 0 RQ9.5 0 0 0 1 0 0 0 0 3 0 0 RQ9.6 0 0 0 0 0 0 0 0 7 0 0 TB 11 3 7 0 0 0 186 0 0 0 0 New New ctenophore New arthropod strata IWMA Ptychoparid Yawunik Stanleycaris Tokummia Priapulid Zacanthoides Kootenayscolex arthropod M A J WQ+120 0 0 0 0 0 0 0 0 0 0 0 WQ0 0 0 0 0 0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 0 0 0 0 0 WQ120 0 0 0 0 0 0 0 0 1 0 0 WQ130 0 0 0 0 0 0 0 0 0 0 0 WQ150 0 0 0 0 0 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 0 0 0 0 0 WQ210 0 0 0 0 0 0 0 0 0 0 0 WQ235 0 0 0 0 0 0 0 0 1 0 0 WQ245 0 0 0 0 0 0 0 0 0 0 0 WQ250 0 0 0 0 0 0 0 0 0 0 0 WQ260 0 0 0 0 0 0 0 0 4 0 0 WQ265 0 0 0 0 0 0 0 0 0 0 0 WQ310 0 0 0 0 0 0 0 0 0 0 0 WQ320 0 0 0 0 0 0 0 0 0 0 0 WQ370 0 0 0 0 0 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 0 2 0 0 WQ400 0 0 0 0 0 0 0 0 3 0 0 WQ420 0 0 0 0 0 0 0 0 2 0 0 WQ430 0 0 0 0 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 0 3 0 0 WQ465 0 0 0 0 0 0 0 0 0 0 0 WQ480 0 0 0 0 0 0 0 0 0 0 0 MC230 0 0 0 0 0 0 0 0 1 1 9 MC240 1 0 0 0 1 0 0 0 0 1 11 MC300 0 1 1 0 0 0 0 0 1 0 19 MC310 1 1 1 0 1 0 0 0 0 0 6 MC320 1 1 0 0 0 0 1 0 2 0 1 MC330 4 2 2 0 4 0 0 0 5 0 9 MC340 8 0 10 0 2 0 0 0 14 0 38 MC350 18 4 24 0 9 0 0 0 29 1 50 MC360 2 0 10 0 1 0 0 0 10 1 8 MC370 8 3 5 0 8 0 1 0 46 0 91 MC380 6 3 48 0 11 0 1 0 224 7 241 MC390 6 0 17 0 9 0 0 0 160 0 57 MC400 8 4 22 0 23 0 3 0 194 0 60 MC410 4 4 18 0 7 0 8 0 109 2 28 MC420 2 0 7 0 5 1 1 0 23 1 2 MC480 1 53 2 0 3 0 0 1 1 0 0 MC490 1 51 1 0 6 0 0 0 0 1 1 MC500 0 4 0 0 4 0 0 0 0 0 5 234

RQ8.0 0 0 0 0 0 0 0 0 0 0 0 RQ8.2 0 1 0 0 0 0 0 0 0 0 0 RQ8.3 0 0 0 1 0 0 0 0 0 0 0 RQ8.4 0 0 0 0 0 0 0 0 0 0 0 RQ8.5 0 0 0 0 0 0 0 0 0 0 0 RQ8.6 0 0 0 0 0 0 0 0 0 0 0 RQ8.8 0 0 0 0 0 0 0 0 0 0 0 RQ8.9 0 0 0 0 0 0 0 0 0 0 0 RQ9.0 0 0 0 0 0 0 0 0 0 0 0 RQ9.1 0 0 0 0 0 0 0 0 0 0 0 RQ9.2 0 0 0 0 0 0 0 0 0 0 0 RQ9.3 0 0 0 0 0 0 0 0 0 0 0 RQ9.4 0 0 0 0 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 0 0 0 0 0 0 RQ9.6 0 0 0 0 0 0 0 0 0 0 0 TB 0 0 0 0 0 0 1 0 0 0 0 New Indet New dinocariidid New New dinocariidid New hemichordate strata Itagnostus Linnarssonia Nectocaris hemichordate A algae A arthropod H D C WQ+120 0 0 0 0 0 0 0 0 0 WQ0 0 0 0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 0 0 0 WQ120 0 0 0 0 0 0 0 0 0 WQ130 0 0 0 0 0 0 0 0 0 WQ150 0 0 0 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 0 0 0 WQ210 0 0 0 0 0 0 0 0 0 WQ235 0 0 0 0 0 0 0 0 0 WQ245 0 0 0 0 0 0 0 0 0 WQ250 0 0 0 0 0 0 0 0 0 WQ260 0 0 0 0 0 0 0 0 0 WQ265 0 0 0 0 0 0 0 0 0 WQ310 0 0 0 0 0 0 0 0 0 WQ320 0 0 0 0 0 0 0 0 0 WQ370 0 0 0 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 0 0 WQ400 0 0 0 0 0 0 0 0 0 WQ420 0 0 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 0 0 WQ480 0 0 0 0 0 0 0 0 0 MC230 0 0 0 0 0 0 0 0 0 MC240 1 1 0 0 0 0 0 0 0 MC300 1 0 0 0 1 0 0 0 0 MC310 1 0 0 0 0 0 0 0 0 MC320 0 0 0 1 0 0 0 0 0 MC330 0 0 0 0 0 1 0 0 0 MC340 0 0 0 1 0 0 1 0 0 MC350 1 0 0 1 1 0 0 1 1 MC360 0 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 0 MC380 0 1 3 2 0 0 4 1 0 MC390 0 2 0 0 0 0 1 1 0 MC400 0 4 0 4 1 2 2 0 1 MC410 0 1 2 3 0 0 0 0 1 235

MC420 0 0 0 1 0 0 0 0 0 MC480 0 4 0 0 0 0 0 1 0 MC490 8 3 0 0 0 0 0 0 0 MC500 2 0 0 0 0 0 0 0 1 RQ8.0 0 0 0 0 0 0 0 1 0 RQ8.2 0 0 0 0 0 0 0 0 0 RQ8.3 0 0 0 0 0 0 0 0 0 RQ8.4 0 0 0 0 0 2 0 0 0 RQ8.5 0 0 0 0 0 0 0 0 0 RQ8.6 0 0 0 0 0 0 0 0 0 RQ8.8 0 0 0 0 0 0 0 0 0 RQ8.9 0 0 0 0 0 0 0 1 0 RQ9.0 0 0 0 0 0 1 0 1 0 RQ9.1 0 0 0 0 0 0 0 0 0 RQ9.2 0 0 0 0 0 0 0 0 0 RQ9.3 0 0 0 0 0 0 0 0 0 RQ9.4 0 0 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 1 0 0 0 RQ9.6 0 0 0 0 0 0 0 0 0 TB 0 0 0 0 0 0 0 0 0 New Chaetognath New Arthropod New ctenophore New arthropod New arthropod strata Surusicaris Banffia Primicaris dinocariidid C spines E B E C WQ+120 0 0 0 0 0 0 0 0 0 WQ0 0 0 0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 0 0 0 WQ120 0 0 0 0 0 0 0 0 0 WQ130 0 0 0 0 0 0 0 0 0 WQ150 0 0 0 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 0 0 0 WQ210 0 0 0 0 0 0 0 0 0 WQ235 0 0 0 0 0 0 0 0 0 WQ245 0 0 0 0 0 0 0 0 0 WQ250 0 0 0 0 0 0 0 0 0 WQ260 0 0 0 0 0 0 0 0 0 WQ265 0 0 0 0 0 0 0 0 0 WQ310 0 0 0 0 0 0 0 0 0 WQ320 0 0 0 0 0 0 0 0 0 WQ370 0 0 0 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 0 0 WQ400 0 0 0 0 0 0 0 0 0 WQ420 0 0 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 0 0 WQ480 0 0 0 0 0 0 0 0 0 MC230 0 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 0 MC350 1 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 0 MC370 0 1 0 0 0 0 0 0 0 236

MC380 0 0 1 4 1 2 1 1 1 MC390 0 0 0 0 0 0 0 0 75 MC400 0 0 1 1 1 0 0 0 26 MC410 0 0 0 0 0 0 0 0 17 MC420 0 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 0 RQ8.0 0 0 0 0 0 0 14 0 0 RQ8.2 0 0 0 0 0 0 0 0 0 RQ8.3 0 0 0 0 0 0 0 0 0 RQ8.4 0 0 0 0 0 0 0 0 0 RQ8.5 0 0 0 0 0 0 0 0 0 RQ8.6 0 0 0 0 0 0 0 0 0 RQ8.8 0 0 0 0 0 0 0 0 0 RQ8.9 0 0 0 0 0 0 0 0 0 RQ9.0 0 0 0 0 0 0 0 0 0 RQ9.1 0 0 0 0 0 0 0 0 0 RQ9.2 0 0 0 0 0 0 0 0 0 RQ9.3 0 0 0 0 0 0 0 0 0 RQ9.4 0 0 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 0 0 0 0 RQ9.6 0 0 0 0 0 0 0 0 0 TB 0 0 0 0 0 0 0 0 0 New New New Hemichordate strata New polychaete B Amplectobelua New sponge B New lobopod B New sponge A polychaete A Arthropod H lobopod A tubes WQ+120 0 0 0 0 0 0 0 0 0 WQ0 0 0 0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 0 0 0 WQ120 0 0 0 0 0 0 0 0 0 WQ130 0 0 0 0 0 0 0 0 0 WQ150 0 0 0 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 0 0 0 WQ210 0 0 0 0 0 0 0 0 0 WQ235 0 0 0 0 0 0 0 0 0 WQ245 0 0 0 0 0 0 0 0 0 WQ250 0 0 0 0 0 0 0 0 0 WQ260 0 0 0 0 0 0 0 0 0 WQ265 0 0 0 0 0 0 0 0 0 WQ310 0 0 0 0 0 0 0 0 0 WQ320 0 0 0 0 0 0 0 0 0 WQ370 0 0 0 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 0 0 WQ400 0 0 0 0 0 0 0 0 0 WQ420 0 0 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 0 0 WQ480 0 0 0 0 0 0 0 0 0 MC230 0 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 0 237

MC340 0 0 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 0 0 MC390 20 2 0 0 0 0 0 0 0 MC400 0 1 1 1 1 0 0 0 0 MC410 0 0 0 0 0 1 1 0 0 MC420 0 0 0 0 0 0 0 1 1 MC480 0 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 1 0 0 RQ8.0 0 0 0 0 0 0 0 0 0 RQ8.2 0 0 0 0 0 0 0 0 0 RQ8.3 0 0 0 0 0 0 0 0 0 RQ8.4 0 0 0 0 0 0 0 0 0 RQ8.5 0 0 0 0 0 0 0 0 0 RQ8.6 0 0 0 0 0 0 0 0 0 RQ8.8 0 0 0 0 0 0 0 0 0 RQ8.9 0 0 0 0 0 0 0 0 0 RQ9.0 0 0 0 0 0 0 0 0 0 RQ9.1 0 0 0 0 0 0 0 0 0 RQ9.2 0 0 0 0 0 0 0 0 0 RQ9.3 0 0 0 0 0 0 0 0 0 RQ9.4 0 0 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 0 0 0 0 RQ9.6 0 0 0 0 0 0 0 0 0 TB 0 0 0 5 0 0 0 0 0 New Tuzoia Naraoia Darth strata Allonia Dictyonina Helcionellid Capinatator Parkaspis Halichondrites Enteropneust RQ-A hemichordate B retifera magna Vader WQ+120 0 0 0 0 0 0 0 0 0 0 0 WQ0 0 0 0 0 0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 0 0 0 0 0 WQ120 0 0 0 0 0 0 0 0 0 0 0 WQ130 0 0 0 0 0 0 0 0 0 0 0 WQ150 0 0 0 0 0 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 0 0 0 0 0 WQ210 0 0 0 0 0 0 0 0 0 0 0 WQ235 0 0 0 0 0 0 0 0 0 0 0 WQ245 0 0 0 0 0 0 0 0 0 0 0 WQ250 0 0 0 0 0 0 0 0 0 0 0 WQ260 0 0 0 0 0 0 0 0 0 0 0 WQ265 0 0 0 0 0 0 0 0 0 0 0 WQ310 0 0 0 0 0 0 0 0 0 0 0 WQ320 0 0 0 0 0 0 0 0 0 0 0 WQ370 0 0 0 0 0 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 0 0 0 0 WQ400 0 0 0 0 0 0 0 0 0 0 0 WQ420 0 0 0 0 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 0 0 0 0 WQ480 0 0 0 0 0 0 0 0 0 0 0 MC230 0 0 0 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 0 0 0 238

MC300 0 0 0 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 0 0 0 0 MC410 0 0 0 0 0 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 0 0 0 MC480 2 0 0 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 0 0 0 RQ8.0 0 4 0 0 1 0 0 0 0 0 12 RQ8.2 0 36 0 0 1 0 3 5 1 0 4 RQ8.3 0 37 0 1 0 0 0 1 0 0 8 RQ8.4 0 200 0 0 2 0 0 3 3 0 51 RQ8.5 0 34 0 0 1 0 0 3 1 0 17 RQ8.6 0 4 0 0 1 0 0 0 0 0 12 RQ8.8 0 13 0 0 1 0 0 0 1 1 4 RQ8.9 0 35 0 0 4 0 0 0 1 0 2 RQ9.0 0 68 0 0 3 0 0 0 1 0 0 RQ9.1 0 15 2 0 2 1 0 0 1 0 1 RQ9.2 0 19 0 0 2 1 0 0 1 0 1 RQ9.3 0 3 0 0 0 0 0 0 1 0 1 RQ9.4 0 0 1 0 0 1 0 0 0 0 0 RQ9.5 0 4 0 0 1 0 2 0 0 0 0 RQ9.6 0 12 1 0 3 0 0 0 2 0 5 TB 0 0 0 12 0 0 0 0 0 0 0 Enteropneus Indet New dinocariidid strata Thelxiope Protoprisma Medusoid Polychaeta RQ-A Archiasterella Dinocariidid RQ-A t RQ-B echinoderm RQ-B WQ+12 0 0 0 0 0 0 0 0 0 0 WQ0 0 0 0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 0 0 0 WQ120 0 0 0 0 0 0 0 0 0 WQ130 0 0 0 0 0 0 0 0 0 WQ150 0 0 0 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 0 0 0 WQ210 0 0 0 0 0 0 0 0 0 WQ235 0 0 0 0 0 0 0 0 0 WQ245 0 0 0 0 0 0 0 0 0 WQ250 0 0 0 0 0 0 0 0 0 WQ260 0 0 0 0 0 0 0 0 0 WQ265 0 0 0 0 0 0 0 0 0 WQ310 0 0 0 0 0 0 0 0 0 WQ320 0 0 0 0 0 0 0 0 0 WQ370 0 0 0 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 0 0 WQ400 0 0 0 0 0 0 0 0 0 WQ420 0 0 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 0 0 239

WQ455 0 0 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 0 0 WQ480 0 0 0 0 0 0 0 0 0 MC230 0 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 0 0 MC410 0 0 0 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 0 RQ8.0 0 0 0 1 0 4 0 0 0 RQ8.2 0 0 3 3 1 1 0 0 3 RQ8.3 0 0 0 2 0 8 0 0 3 RQ8.4 0 0 0 2 2 30 0 3 0 RQ8.5 0 2 2 1 0 5 0 0 0 RQ8.6 0 0 0 1 0 10 0 0 0 RQ8.8 5 0 0 2 0 1 0 0 1 RQ8.9 1 0 0 0 0 2 0 0 1 RQ9.0 0 0 0 5 0 5 0 3 1 RQ9.1 0 0 0 1 0 19 1 0 0 RQ9.2 0 0 0 1 0 4 1 0 0 RQ9.3 0 0 0 1 0 0 0 0 0 RQ9.4 0 0 0 0 0 0 0 0 1 RQ9.5 0 0 0 10 0 0 0 0 0 RQ9.6 0 0 0 6 0 1 0 0 0 TB 0 0 0 0 0 0 0 0 0 New arthropod Proboscicari Leanchoilia Anoria Byronia Nereocaris Ogygopsis strata Alokistocare RQ-A s persephone sp. annulata excilis klotzi WQ+120 0 0 0 0 0 0 0 0 WQ0 0 0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 0 0 WQ120 0 0 0 0 0 0 0 0 WQ130 0 0 0 0 0 0 0 0 WQ150 0 0 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 0 0 WQ210 0 0 0 0 0 0 0 0 WQ235 0 0 0 0 0 0 0 0 WQ245 0 0 0 0 0 0 0 0 WQ250 0 0 0 0 0 0 0 0 WQ260 0 0 0 0 0 0 0 0 WQ265 0 0 0 0 0 0 0 0 WQ310 0 0 0 0 0 0 0 0 WQ320 0 0 0 0 0 0 0 0 WQ370 0 0 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 0 240

WQ400 0 0 0 0 0 0 0 0 WQ420 0 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 0 WQ480 0 0 0 0 0 0 0 0 MC230 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 0 MC410 0 0 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 RQ8.0 0 0 0 0 0 0 0 0 RQ8.2 2 1 2 0 0 0 0 0 RQ8.3 0 0 0 0 0 0 0 0 RQ8.4 0 0 1 0 0 0 0 0 RQ8.5 0 0 0 0 0 0 0 0 RQ8.6 0 0 2 0 0 0 0 0 RQ8.8 0 0 0 0 0 0 0 0 RQ8.9 0 0 4 0 0 0 0 0 RQ9.0 0 2 3 0 0 0 0 0 RQ9.1 0 1 2 0 0 0 0 0 RQ9.2 0 1 2 0 0 0 0 0 RQ9.3 0 0 4 0 0 0 0 0 RQ9.4 0 0 2 0 0 0 0 0 RQ9.5 0 2 0 0 0 0 0 0 RQ9.6 0 1 1 0 0 0 0 0 TB 0 0 0 1 1 4 3 9 Oikozetetes Oryctocara Leptomitella Hintzespongia Leanchoilia Laeanigma strata Glossopleura Poliela prima seilacheri geikiei incepta sp. protagonia striatum WQ+120 0 0 0 0 0 0 0 0 WQ0 0 0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 0 0 WQ120 0 0 0 0 0 0 0 0 WQ130 0 0 0 0 0 0 0 0 WQ150 0 0 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 0 0 WQ210 0 0 0 0 0 0 0 0 WQ235 0 0 0 0 0 0 0 0 WQ245 0 0 0 0 0 0 0 0 WQ250 0 0 0 0 0 0 0 0 WQ260 0 0 0 0 0 0 0 0 WQ265 0 0 0 0 0 0 0 0 241

WQ310 0 0 0 0 0 0 0 0 WQ320 0 0 0 0 0 0 0 0 WQ370 0 0 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 0 WQ400 0 0 0 0 0 0 0 0 WQ420 0 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 0 WQ480 0 0 0 0 0 0 0 0 MC230 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 0 MC410 0 0 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 RQ8.0 0 0 0 0 0 0 0 0 RQ8.2 0 0 0 0 0 0 0 0 RQ8.3 0 0 0 0 0 0 0 0 RQ8.4 0 0 0 0 0 0 0 0 RQ8.5 0 0 0 0 0 0 0 0 RQ8.6 0 0 0 0 0 0 0 0 RQ8.8 0 0 0 0 0 0 0 0 RQ8.9 0 0 0 0 0 0 0 0 RQ9.0 0 0 0 0 0 0 0 0 RQ9.1 0 0 0 0 0 0 0 0 RQ9.2 0 0 0 0 0 0 0 0 RQ9.3 0 0 0 0 0 0 0 0 RQ9.4 0 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 0 0 0 RQ9.6 0 0 0 0 0 0 0 0 TB 47 1 3 8 1 3 156 1 Polypearaspi Sanshapentalla Siphusauctum strata Undet Algae TB-A Undet Algae TB-B Undet Cyanobacteria TB-A Undet Cyanobacteria TB-B s sp. sp. gregarium WQ+120 0 0 0 0 0 0 0 WQ0 0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 0 WQ120 0 0 0 0 0 0 0 WQ130 0 0 0 0 0 0 0 WQ150 0 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 0 WQ210 0 0 0 0 0 0 0 WQ235 0 0 0 0 0 0 0 242

WQ245 0 0 0 0 0 0 0 WQ250 0 0 0 0 0 0 0 WQ260 0 0 0 0 0 0 0 WQ265 0 0 0 0 0 0 0 WQ310 0 0 0 0 0 0 0 WQ320 0 0 0 0 0 0 0 WQ370 0 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 WQ400 0 0 0 0 0 0 0 WQ420 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 WQ480 0 0 0 0 0 0 0 MC230 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 MC410 0 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 RQ8.0 0 0 0 0 0 0 0 RQ8.2 0 0 0 0 0 0 0 RQ8.3 0 0 0 0 0 0 0 RQ8.4 0 0 0 0 0 0 0 RQ8.5 0 0 0 0 0 0 0 RQ8.6 0 0 0 0 0 0 0 RQ8.8 0 0 0 0 0 0 0 RQ8.9 0 0 0 0 0 0 0 RQ9.0 0 0 0 0 0 0 0 RQ9.1 0 0 0 0 0 0 0 RQ9.2 0 0 0 0 0 0 0 RQ9.3 0 0 0 0 0 0 0 RQ9.4 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 0 0 RQ9.6 0 0 0 0 0 0 0 TB 1 1 1525 32 28 17 5 Undet Lobopodian TB- strata Undet dinocaridiid TB-A Undet eocrinoid TB-A Undet Lobopodia TB-B Undet Porifera TB-A Undet Porifera TB-B A WQ+120 0 0 0 0 0 0 WQ0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 WQ120 0 0 0 0 0 0 WQ130 0 0 0 0 0 0 243

WQ150 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 WQ210 0 0 0 0 0 0 WQ235 0 0 0 0 0 0 WQ245 0 0 0 0 0 0 WQ250 0 0 0 0 0 0 WQ260 0 0 0 0 0 0 WQ265 0 0 0 0 0 0 WQ310 0 0 0 0 0 0 WQ320 0 0 0 0 0 0 WQ370 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 WQ400 0 0 0 0 0 0 WQ420 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 WQ480 0 0 0 0 0 0 MC230 0 0 0 0 0 0 MC240 0 0 0 0 0 0 MC300 0 0 0 0 0 0 MC310 0 0 0 0 0 0 MC320 0 0 0 0 0 0 MC330 0 0 0 0 0 0 MC340 0 0 0 0 0 0 MC350 0 0 0 0 0 0 MC360 0 0 0 0 0 0 MC370 0 0 0 0 0 0 MC380 0 0 0 0 0 0 MC390 0 0 0 0 0 0 MC400 0 0 0 0 0 0 MC410 0 0 0 0 0 0 MC420 0 0 0 0 0 0 MC480 0 0 0 0 0 0 MC490 0 0 0 0 0 0 MC500 0 0 0 0 0 0 RQ8.0 0 0 0 0 0 0 RQ8.2 0 0 0 0 0 0 RQ8.3 0 0 0 0 0 0 RQ8.4 0 0 0 0 0 0 RQ8.5 0 0 0 0 0 0 RQ8.6 0 0 0 0 0 0 RQ8.8 0 0 0 0 0 0 RQ8.9 0 0 0 0 0 0 RQ9.0 0 0 0 0 0 0 RQ9.1 0 0 0 0 0 0 RQ9.2 0 0 0 0 0 0 RQ9.3 0 0 0 0 0 0 RQ9.4 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 0 RQ9.6 0 0 0 0 0 0 TB 1 30 36 4 1 163 Undet priapulid Undet strata Undet Taxon TB-A Undet Taxon TB-B Undet Taxon TB-C Undet Taxon TB-D Undet Taxon TB-E Undet Taxon TB-F TB-A Taxon A WQ+120 0 0 0 0 0 0 0 0 244

WQ0 0 0 0 0 0 0 0 0 WQ110 0 0 0 0 0 0 0 0 WQ120 0 0 0 0 0 0 0 0 WQ130 0 0 0 0 0 0 0 0 WQ150 0 0 0 0 0 0 0 0 WQ180 0 0 0 0 0 0 0 0 WQ210 0 0 0 0 0 0 0 0 WQ235 0 0 0 0 0 0 0 0 WQ245 0 0 0 0 0 0 0 0 WQ250 0 0 0 0 0 0 0 0 WQ260 0 0 0 0 0 0 0 0 WQ265 0 0 0 0 0 0 0 0 WQ310 0 0 0 0 0 0 0 0 WQ320 0 0 0 0 0 0 0 0 WQ370 0 0 0 0 0 0 0 0 WQ380 0 0 0 0 0 0 0 0 WQ400 0 0 0 0 0 0 0 0 WQ420 0 0 0 0 0 0 0 0 WQ430 0 0 0 0 0 0 0 0 WQ445 0 0 0 0 0 0 0 0 WQ455 0 0 0 0 0 0 0 0 WQ465 0 0 0 0 0 0 0 0 WQ480 0 0 0 0 0 0 0 0 MC230 0 0 0 0 0 0 0 0 MC240 0 0 0 0 0 0 0 0 MC300 0 0 0 0 0 0 0 0 MC310 0 0 0 0 0 0 0 0 MC320 0 0 0 0 0 0 0 0 MC330 0 0 0 0 0 0 0 0 MC340 0 0 0 0 0 0 0 0 MC350 0 0 0 0 0 0 0 0 MC360 0 0 0 0 0 0 0 0 MC370 0 0 0 0 0 0 0 0 MC380 0 0 0 0 0 0 0 0 MC390 0 0 0 0 0 0 0 0 MC400 0 0 0 0 0 0 0 0 MC410 0 0 0 0 0 0 0 0 MC420 0 0 0 0 0 0 0 0 MC480 0 0 0 0 0 0 0 0 MC490 0 0 0 0 0 0 0 0 MC500 0 0 0 0 0 0 0 0 RQ8.0 0 0 0 0 0 0 0 0 RQ8.2 0 0 0 0 0 0 0 0 RQ8.3 0 0 0 0 0 0 0 0 RQ8.4 0 0 0 0 0 0 0 0 RQ8.5 0 0 0 0 0 0 0 0 RQ8.6 0 0 0 0 0 0 0 0 RQ8.8 0 0 0 0 0 0 0 0 RQ8.9 0 0 0 0 0 0 0 0 RQ9.0 0 0 0 0 0 0 0 0 RQ9.1 0 0 0 0 0 0 0 0 RQ9.2 0 0 0 0 0 0 0 0 RQ9.3 0 0 0 0 0 0 0 0 RQ9.4 0 0 0 0 0 0 0 0 RQ9.5 0 0 0 0 0 0 0 0 RQ9.6 0 0 0 0 0 0 0 0 245

TB 5 308 1740 170 56 24 38 63 strata Undet Taxon TB-G Undet Taxon TB-H Undet Worm TB-A WQ+120 0 0 0 WQ0 0 0 0 WQ110 0 0 0 WQ120 0 0 0 WQ130 0 0 0 WQ150 0 0 0 WQ180 0 0 0 WQ210 0 0 0 WQ235 0 0 0 WQ245 0 0 0 WQ250 0 0 0 WQ260 0 0 0 WQ265 0 0 0 WQ310 0 0 0 WQ320 0 0 0 WQ370 0 0 0 WQ380 0 0 0 WQ400 0 0 0 WQ420 0 0 0 WQ430 0 0 0 WQ445 0 0 0 WQ455 0 0 0 WQ465 0 0 0 WQ480 0 0 0 MC230 0 0 0 MC240 0 0 0 MC300 0 0 0 MC310 0 0 0 MC320 0 0 0 MC330 0 0 0 MC340 0 0 0 MC350 0 0 0 MC360 0 0 0 MC370 0 0 0 MC380 0 0 0 MC390 0 0 0 MC400 0 0 0 MC410 0 0 0 MC420 0 0 0 MC480 0 0 0 MC490 0 0 0 MC500 0 0 0 RQ8.0 0 0 0 RQ8.2 0 0 0 RQ8.3 0 0 0 RQ8.4 0 0 0 RQ8.5 0 0 0 RQ8.6 0 0 0 RQ8.8 0 0 0 RQ8.9 0 0 0 RQ9.0 0 0 0 RQ9.1 0 0 0 RQ9.2 0 0 0 246

RQ9.3 0 0 0 RQ9.4 0 0 0 RQ9.5 0 0 0 RQ9.6 0 0 0 TB 3 8 129

247

Appendix 2. Burgess Shale species composition data matrix.

248

Locali PESU PEHS ESSU NKHS IVHS EVHS EPP NKDE EVDE EVGR ESDE ESHS NKSU ty TB 8 411 2699 286 78 155 1029 295 93 20 0 85 1 MC23 0 2 258 27 0 2 3 27 1 0 0 0 2 0 MC24 0 1 416 45 0 2 10 42 0 0 0 0 6 0 MC30 0 0 433 48 0 1 2 11 1 0 0 0 0 0 MC31 0 0 239 19 0 1 2 28 0 0 0 0 0 0 MC32 0 0 273 30 0 2 1 32 2 0 0 0 0 0 MC33 2 1 271 83 0 2 0 100 5 0 0 0 5 0 MC34 2 2 255 68 1 11 1 51 14 0 0 0 3 0 MC35 2 2 568 328 1 27 3 73 29 0 0 0 5 0 MC36 3 3 197 206 2 10 0 62 10 0 0 0 11 0 MC37 0 1 659 493 1 8 6 94 46 0 0 0 29 0 MC38 6 16 703 934 1 56 6 257 224 0 0 0 79 0 MC39 4 1 265 377 0 17 3 96 160 2 0 0 9 0 MC40 2 3 816 958 1 25 0 241 194 0 0 0 13 0 MC41 9 2 418 1358 0 21 5 183 109 0 0 0 26 0 MC42 1 1 70 464 1 7 20 47 23 0 0 0 4 0 MC48 0 0 38 16 0 5 0 248 1 0 0 0 18 0 MC49 0 2 22 11 0 2 2 396 0 0 0 0 10 0 MC50 0 0 4 13 0 2 2 467 0 0 0 0 8 0 WQ+1 0 11 39 33 50 9 54 134 122 15 0 1 0 20 WQ0 3 13 45 122 59 3 231 650 191 53 0 0 0 WQ11 0 11 66 165 11 16 9 72 206 12 0 0 0 0 WQ12 10 34 291 813 328 40 463 544 395 61 0 1 0 0 WQ13 12 8 197 190 55 32 204 1697 429 251 0 95 1 0 WQ15 11 51 670 389 37 19 127 987 198 168 0 31 0 0 WQ17 8 29 524 95 60 24 121 450 28 46 0 4 0 0 WQ21 18 44 1264 470 228 117 310 1354 227 131 2 15 5 0 WQ23 23 42 593 405 168 117 220 375 111 80 0 4 0 5 WQ24 2 63 748 1413 304 1 188 740 746 126 0 28 0 5 WQ25 3 7 422 884 328 4 66 179 435 40 2 7 0 0 WQ26 96 37 553 1155 169 10 62 440 213 221 0 0 0 0 WQ26 12 18 396 453 78 9 44 210 46 57 0 2 0 5 WQ31 8 8 202 149 22 12 32 159 26 41 0 2 0 0

249

WQ32 10 7 557 136 46 39 30 499 186 32 0 0 1 0 WQ37 2 7 242 35 12 0 22 140 58 27 0 0 0 0 WQ38 1 7 55 75 29 2 66 75 69 7 0 0 0 0 WQ40 9 30 443 151 139 54 792 375 325 93 0 0 4 0 WQ42 1 2 170 624 34 0 9 269 347 69 0 0 0 0 WQ43 0 3 57 71 5 1 0 63 110 25 0 0 0 0 WQ44 0 6 1093 101 9 1 8 109 97 9 0 0 0 5 WQ45 0 7 17 21 2 10 2 124 37 38 0 0 0 5 WQ46 0 1 232 16 2 0 1 19 396 11 0 0 0 5 WQ48 23 3 59 41 12 11 16 155 29 13 0 0 2 0 RQ8.0 5 16 139 115 22 0 0 7 0 1 0 0 0 RQ8.2 5 33 346 194 32 0 4 1 2 1 0 0 0 RQ8.3 10 25 186 126 79 0 5 0 0 0 0 0 0 RQ8.4 3 28 955 172 58 0 22 10 1 2 0 0 0 RQ8.5 2 22 338 130 27 4 26 5 0 1 0 0 0 RQ8.6 7 13 205 80 31 0 6 4 2 1 0 0 0 RQ8.8 10 27 198 196 74 0 12 5 2 1 0 0 0 RQ8.9 2 13 213 133 43 0 15 5 1 4 0 0 0 RQ9.0 1 32 417 164 86 0 12 11 2 4 0 0 0 RQ9.1 6 15 152 128 58 2 16 1 1 3 0 0 0 RQ9.2 4 18 188 173 70 2 24 0 1 3 0 0 0 RQ9.3 4 19 292 194 125 0 15 0 0 0 1 0 0 RQ9.4 3 29 224 122 51 1 8 0 0 0 0 0 0 RQ9.5 4 17 91 130 56 0 3 1 0 1 0 0 0 RQ9.6 5 13 172 283 169 0 8 1 0 3 0 0 0

250

Appendix 3. Burgess Shale ecological composition data matrix.

251

GENERAL CONCLUSIONS

THE UTILITY OF TAPHONOMIC AND SPECIES-LEVEL STUDIES FOR PALEOCOMMUNITY RECONSTRUCTIONS

“One creature is tantalizing; but we need whole faunas for sound conclusions.”

Stephen Jay Gould, 1989

Reconstructing the diversity and niche structure of any paleocommunity is a significant undertaking, as the phylogenetic and ecological affinities of the constituent taxa are often cryptic and taphonomic bias often influences patterns of species distribution and abundance. Marble

Canyon, where nearly one quarter of all species are currently hypothesized to be new, illustrates these points clearly. The first three chapters of my dissertation are intended to mitigate these issues while simultaneously addressing their own major taphonomic and evolutionary questions.

Chapter 1 focused on the sequence of decay of morphological characters of three species of acorn worm (Class: Enteropneusta, Phylum: Hemichordata). The broad goal of this chapter was to provide a framework for the interpretation of hemichordate fossils with reference to the morphology of extant forms, while accounting for possible taphonomic biases. More crucially for the paleocommunity reconstruction, however, was the utility of this dataset for assessing possible taphonomic biases at the level of bedding assemblages using Oesia disjuncta as an index species. As has been done using the timeline of polychaete decay (Briggs and Kear 1993) to assess taphonomic bias at the Walcott Quarry (Caron and Jackson 2006), we were able to discern from our decay data that the presence of pristine Oesia specimens across nearly all bedding assemblages suggests that pre- and post-depositional transport was not a significant confounding variable in describing the paleocommunity at Marble Canyon. It also suggests that the wholesale loss of entire specimens to post-mortem decay was minimal, as the timeline of

252 hemichordate decay is among the fastest thus far described (Sansom et al. 2010, 2013; Murdock et al. 2014).

Chapter 2 redescribed Oesia disjuncta as a hemichordate and Margaretia dorus as its tube-shaped dwelling. The importance of this chapter cannot be understated for the reconstruction of the Marble Canyon paleocommunity. O. disjuncta (3,373 specimens) and M. dorus (1,312 specimens) are the third and fifth most abundant fossils at Marble Canyon, respectively, making a clear understanding of the affinities of these fossils crucial to interpreting the structure and diversity of the paleocommunity. (M. dorus, now known to be the tube of

Oesia, is not included in Chapter 4 as a valid taxon, thus the position of fifth most abundant taxon goes to Kootenayscolex barbarensis).

Had Margaretia remained a large, tubular alga, our picture of the Marble Canyon environment would likely have been that of a relatively shallow-water community due to its great abundance. Furthermore, the description of Oesia as a tube-building hemichordate impacts the community analysis significantly from both a taxonomic and ecological perspective. In terms of representative major taxonomic groups, Marble Canyon is the most hemichordate-rich

Burgess Shale locality, suggesting—in tandem with the abundance of the recently described hemichordate Spartobranchus tenuis (Caron et al. 2013)—that this phylum was far more significant in early paleoecology than previously thought. This view is made even more significant due the wide geographic range of Margaretia, including sites in China and Russia

(Nanglu et al. 2016). Finally, the sessile-suspension-feeding niche we now infer for Oesia differs significantly from previous hypotheses. Had this study not been undertaken, Oesia may have been attributed to either a planktonic-predatory (Szaniawski 2005), or a motile-deposit feeding niche (Walcott 1911).

253

Finally, Chapter 3 described the first new annelid (Kootenayscolex barbarensis) from the

Burgess Shale in nearly 40 years (Morris 1979). While the principal significance of this discovery are its evolutionary ramifications (Nanglu and Caron 2018), an in-depth study of this taxon was necessary to distinguish it from Burgessochaeta, the taxon to which it was originally assigned (Caron et al. 2014). As Kootenayscolex is the fifth most abundant taxon in our analyses, the fact that it is correctly identified as a different species from those that are primarily known from the Walcott Quarry was critical to distinguishing the two faunas quantitatively.

Kootenayscolex is also the preeminent epibenthic-vagrant-deposit feeder at Marble Canyon, a mode of life that could only be inferred by observations of its pharyngeal and gut contents

(Nanglu and Caron 2018).

MORPHOLOGICAL INNOVATION, EVO-DEVO AND THE ORIGIN OF BODY PLANS

The Cambrian Explosion is not only significant for the rapid appearance of nearly all major metazoan phyla within a short geologic time span, although that alone is certainly more than enough to warrant the attention it has received (Erwin et al. 2011). Another major component of what distinguishes this phenomenon as one of the most significant biological events to have occurred in the Earth’s history are the unusual morphologies of Cambrian taxa— at least in comparison with modern analogues—which seemed to exist only over a relatively brief period of time. Many of these taxa are so unusual in their appearance that it seemed plausible, for a time, that morphological disparity and phyletic diversity reached their apex in the

Cambrian, inexorably decreasing toward current levels (Conway Morris 1986; Gould 1989).

While the phylogenetic affinities of many fossil taxa from this period remain cryptic, for

254 example, the tulip animal (O’Brien and Caron 2012), or the eldonids and their discoidal kin (Zhu et al. 2002; Caron et al. 2010), the advent and popularization of the stem- vs. crown-group paradigm for describing fossil taxa combined with more sophisticated methods, particularly molecular data, have disproven this notion (at least as far as the total number of phyla are concerned) (Jensen and Budd 2000; Briggs and Fortey 2005). However, the phylogenetic positioning of most of these taxa as stem-group members of major modern animal groups has not lessened their interest or significance. Indeed, it is because of this phylogenetic positioning that the Cambrian biota is unequalled in illustrating the origins of phyla and their body plans.

To that end, plasticity in, or at least more limited constraints on, gene regulatory networks has often been invoked as a mechanism accounting for the unique body plans found in the Cambrian (Davidson and Erwin 2006; Peterson et al. 2009; Erwin et al. 2011), and as a mechanism for body plan evolution in general (Shubin et al. 2009; Peter and Davidson 2011).

Throughout my dissertation, I have come to appreciate the explanatory power that ontogenetic studies and evolutionary developmental biology can provide for reconciling extant taxa with their half-billion-year old cousins. The head morphology of Kootenayscolex (Chapter 3) presents an interesting dilemma in that the apparent scale of the morphological conundrum in question belies its significance (Nanglu and Caron 2018). That conundrum was, in essence, a single anomalous character: finding parapodia and chaetae surrounding the mouth. How did these structures, patterned at the embryological level as exclusively structures of the body segments, end up on the head? Our solution as it stands currently—I have no doubt that further investigation of the Cambrian annelid fossil record will refine our theories—was to draw parallels with the developmental biology of extant polychaetes such as nereidids, which fuse body segments with the head, and the magelonids, which selectively re-absorb parapodia and

255 chaetae during ontogeny. The observation that both Canadia spinosa and Burgessochaeta setigera also appear to possess a similar organization of head appendages also seems promising for expanding on the utility of extant developmental pathways to provide insight into Cambrian morphology.

More interesting from a personal perspective, however, are the ambulacrarians. The early evolutionary history of the Ambulacraria (Hemichordata+Echinodermata) remains extremely contentious, due in large part to the high degree of morphological disparity between its constituent phyla. However, recent fossil discoveries have demonstrated an unexpected wealth of data regarding the form of the earliest ambulacrarians. Perhaps most significant are the vermiform, tube-building hemichordates Spartobranchus tenuis (Caron et al. 2013) and Oesia disjuncta (Nanglu et al. 2016) (Chapter 2), as well as new ambulacrarian-like animals of uncertain affinity such Herpetogaster collinsi (Caron et al. 2010). All three of these taxa possess both unique morphological characters as well as striking combinations of modern anatomical features found among disparate groups, the homologies of which may be elucidated with more robust comparisons with the developmental biology of extant relatives.

Among the most notable characteristics of both S. tenuis and O. disjuncta are the posterior structures that lack a clear analogue among extant hemichordate worms. In S. tenuis, it was suggested that the posterior structures may have served as a form of anchor due to their bulbous but simple shape. The posterior structure of O. disjuncta, which is claw-shaped and presumably muscular based on the style of preservation, was also hypothesized to serve as an attachment structure, although of a more active nature than that of S. tenuis. While the morphology of the adult Enteropneusta—which S. tenuis and O. disjuncta most closely resemble from an anatomical standpoint—does not significantly clarify the origins or homology of this

256 structure, two other comparisons from within Hemichordata may. The first is the observation that harrimaniid enteropneusts use their juvenile post-anal tails as attachment structures that have been hypothesized to be homologous with the pterobranch stolon (Cameron 2002). The second is the observation of bilateral posterior structures in developing juvenile specimens of

Schizocardium (Gonzalez et al. 2017) which bear visual parallels with the grasping posterior structure of O. disjuncta. If these characters are indeed homologous, then the phylogenetic position of both S.tenuis and O.disjuncta becomes of paramount importance for reconstructing the origins of the colonial ecology of the Pterobranchia (including Graptolithina (Mitchell et al.

2013)).

O. disjuncta is also an unusual taxon in that its gill bars are very much like those of their extant enteropneust relatives, with one major difference: they extend throughout the entire trunk, rather than occurring only in the pharynx (Nanglu et al. 2016). When considered in conjunction with the tube-building lifestyle of this animal, O. disjuncta was interpreted as a sessile, suspension-feeding animal. In effect, it is almost a perfect chimera of enteropneust and pterobranch characters. Like an enteropneust, O. disjuncta was solitary and vermiform. Like a pterobranch, it was tube-dwelling and suspension-feeding. These features put a primacy on determining the phylogenetic position of O. disjuncta with greater resolution, as its suite of anatomical characters may significantly inform our framework for the early sequence of character acquisition among the Hemichordata. Was the last common ancestor of the hemichordates tube-building or infaunal? When did the miniaturization of the tube-building pterobranchs occur? Was suspension feeding or deposit feeding the ancestral mode of life for

Hemichordata? All of these are questions that further studies of O. disjuncta and S. tenuis may be able to elucidate.

257

Finally, Herpetogaster collinsi remains a somewhat enigmatic taxon. While likely of an ambulacrarian affinity, robust comparisons with a multitude of both extant and fossil taxa did not result in a more detailed phylogenetic position (Caron et al. 2010). Further observations of extant taxa, including their developmental biology, may suggest a more hemichordate-adjacent positioning. Firstly, although the triangular structures of H. collinsi were discussed as possible gill slit openings (Caron et al. 2010), the utility of this feature as phylogenetically informative was considered minimal due to the aforementioned status of gill slits being a deuterostome plesiomorphy (Lowe et al. 2015). This conclusion did not take into account, however, that the reduction or loss of gill slits may in and of itself be pertinent to this question. More specifically, the cephalodiscid pterobranchs and Stereobalanus (an enteropneust belonging to the

Harrimanidae, the earliest branching Family of the Enteropneusta) have a reduced number of gill slits (two) (Cameron et al. 2000). The pterobranch Rhabdopleura has presumably reduced this feature even further, to the point of lacking gill slits at altogether (Cameron et al. 2000).

Additionally, the pterobranch Atubaria was not considered for gross morphological comparisons with H. collinsi. Like H. collinsi and all other pterobranchs, Atubaria has a long stalk and filter feeding arms attached to a bulbous main body, or trunk (Komai 1949). Unlike all other pterobranchs, but like H. collinsi, Atubaria does not live in a coenoecium. Additionally, presumptive juveniles of Atubaria were found to only have “two rod-like processes” that later grew into six feeding arms in its adult form (Komai 1949). This process may provide interesting parallels for the de novo acquisition of feeding arms in Hemichordata, although the rarity of

Atubaria and relative dearth of data regarding this taxon may prohibit further inference. The paired, tentacled, adult feeding arms of the pterobranch Rhabdopleura may prove to be a more fruitful point of comparison with H. collinsi (Ruppert 2005).

258

The questions broached in the preceding pages suggest that, while the potential for applying ontogenetic data to the origins of any phyla are profound, Ambulacraria in particular may represent a tractable study system for exploration. With major strides having been recently made from both the paleontological (Caron et al. 2010, 2013; Nanglu et al. 2016) and developmental fields (Cameron 2002; Gillis et al. 2012; Röttinger and Lowe 2012; Gonzalez et al. 2017), the ambulacrarians may represent a relatively untapped wellspring of data for investigating the role of developmental mechanisms as the genetic substrate for the Cambrian

Explosion (Davidson and Erwin 2006; Peterson et al. 2009; Peter and Davidson 2011; Peterson and Eernisse 2016).

COMMUNITY ECOLOGY DURING THE CAMBRIAN: CURRENT STATUS AND FUTURE WORK

Data from Chapter 4 demonstrate clear patterns of both locality-specific and landscape- scale compositional heterogeneity throughout the Burgess Shale. Considering other recent paleocommunity analyses (O’Brien and Caron 2015), the view that the Greater Phyllopod Bed fauna of the Walcott Quarry is representative of a cosmopolitan middle Cambrian biota cannot be supported. More importantly, patterns of faunal turnover in the Greater Phyllopod Bed community and their inferred causes (Caron and Jackson 2006, 2008) cannot be generalized to the entirety of the Burgess Shale. Each Cambrian locality should be considered as unique with respect to the forces governing the assembly of its communities over time. The abiotic environment, disturbance, competition, and predation act in concert to structure the diversity and geographic distribution of modern benthic communities (Connell 1961a, b; Paine 1974; Hubbell

2001), with the relative significance of each factor varying heterogeneously over large spatio- temporal scales (Connell 1961a, b). The same is likely true for the middle Cambrian.

259

I suggest that further investigations of how early animal communities were structured and what forces may have controlled their assembly are best explored through two complementary avenues. The first approach, more straightforward in its application but perhaps more difficult in execution, is to add more sites to the types of multivariate analyses that are now de rigeur for paleocommunity analyses. Within the Burgess Shale, new discoveries from near the Marble

Canyon may provide the most promising avenue. The majority of these fossils are, however, talus material. As such, they do not provide the same detailed stratigraphic data that would permit faunal turnover comparisons at the same ecological scale as Walcott Quarry, Raymond

Quarry, or Marble Caynon. Nevertheless, they make up for this deficiency by providing a much broader geographic scale than is currently available to us through other avenues. With the

Walcott Quarry and Raymond Quarry most likely representing an extended time series of the same locality, we currently have only two well-represented sites for broad scale comparisons of faunal turnover. Fossils from the relatively nearby Tokumm Creek could also provide a half- dozen or so new data points to help polarize our gradient of compositional change, representing a significant step toward a “landscape” scale understanding of middle Cambrian biogeography.

It is also worth considering the potential to expand the temporal range of these analyses.

The most logical choice to do so would be to include sites from the ~7-million-year older

Maotianshan Shales (Zhao et al. 2013) and the Guanshan biota (Hu et al. 2010). Paleocommunity analyses of these sites have already been conducted, and initial comparisons between the

Chengjiang and the Burgess Shale have already been made (Zhao et al. 2013; O’Brien and Caron

2015). As such, there is already a wealth of information potentially available to integrate with the community data matrix compiled in this dissertation. Such a combined analysis would provide a

260 view of the changing patterns of biodiversity throughout the Cambrian unparalleled in its temporal and taxonomic scope.

Incorporating taxa from the Moroccan Fezouata locality may be even more interesting

(Van Roy et al. 2015a; Lefebvre et al. 2016; Martin et al. 2016). Sampling from a series of well- preserved Ordovician fossil sites which possess many Cambrian-type animals would expand the view of community ecology across a major geological boundary. Including paleocommunities from this locality would provide the rare opportunity to see how animals such as marellomorphs

(Van Roy et al. 2015a) and dinocariidids (Van Roy et al. 2015b) coexisted with a new blossoming of diversity, and how their ecologies were impacted by the advent of this new biota.

This is to say nothing of the potential insights into early Paleozoic ecology that may be gained from comparing the various Fezouata fossil sites with each other in the same manner as the

Burgess Shale and Chengjiang have been analyzed (Caron and Jackson 2008; Zhao et al. 2013;

O’Brien and Caron 2015).

The second approach would be to refine our understanding of Cambrian functional ecology and apply it to ecologically relevant stratigraphic sub-units like those used in the fourth chapter of this dissertation. Considering the wide disparity of body plans and ecologies represented by the taxa within even one bedding assemblage of a Burgess Shale locality, the functional ecologies or “niches” of these taxa must be coded at a very general level. The three- axis framework of tiering-motility-trophic mode (Bambach et al. 2007) is adequate in this regard, and has become commonplace with these types of data (Caron and Jackson 2008; Zhao et al.

2013; O’Brien and Caron 2015). One advantage of the increasingly widespread application of this framework for paleocommunity studies in addition to its broad, descriptive nature is its utility for making comparisons across disparate studies. However, it is easy to criticize the

261 description of an animal as an “epibenthic-sessile-suspension feeder” as too vague for teasing apart fine-scale community structure. For a sponge, a hemichordate, and a brachiopod to be considered functionally equivalent may be useful for understanding major ecological trends, but fine-scale changes in the structure of bedding assemblages may be lost through such generalizations.

Arthropods seem the logical choice for investigating paleoecological phenomena in greater detail, owing to: i) their considerable diversity (Caron and Jackson 2008; Hu et al. 2010;

Zhao et al. 2013; O’Brien and Caron 2015), ii) their equally considerable morphological disparity (Haug et al. 2012; Yang et al. 2013; Aria et al. 2015; Aria and Caron 2017), and iii) their ubiquity throughout soft-bodied Cambrian Lagerstätte. A study of the Marble Canyon dataset, for example, purely through the lens of changes in the morphospace of arthropod feeding appendages and binned by bedding assemblage may illustrate even finer divisions in niche partitioning than we currently estimate.

Whether at the scale of taphonomic and environmental conditions (Chapter 1), the biology and evolutionary affinities of individual species (Chapters 2 and 3), or entire communities and landscapes (Chapter 4), our knowledge of the Cambrian is excitingly incomplete. It remains to be seen to what degree the discovery of new sites of exceptional preservation will recapitulate the ecological patterns and evolutionary insights described throughout the series of studies presented here. Moving toward a holistic understanding of this period will require both a broadening of the scope of our data and, simultaneously, a refinement of its resolution. The goal of this dissertation was to contribute to both of these efforts, and to hopefully provide a new perspective on the world inhabited by our distant relatives.

262

REFERENCES

Aria, C., and J.-B. Caron. 2017. Burgess Shale fossils illustrate the origin of the mandibulate

body plan. Nature 545:89–92.

Aria, C., J.-B. Caron, and R. Gaines. 2015. A large new leanchoiliid from the Burgess Shale and

the influence of inapplicable states on stem arthropod phylogeny. Palaeontology 58:629–

660.

Bambach, R. K., A. M. Bush, and D. H. Erwin. 2007. Autecology and the filling of ecospace:

key metazoan radiations. Palaeontology 50:1–22.

Briggs, D. E. G., and A. J. Kear. 1993. Decay and preservation of polychaetes: taphonomic

thresholds in soft-bodied organisms. Paleobiology 19:107–135.

Briggs, D. E. G., and R. A. Fortey. 2005. Wonderful strife: systematics, stem groups, and the

phylogenetic signal of the Cambrian radiation. Paleobiology 31:94–112.

Cameron, C. B. 2002. The anatomy, life habits, and later development of a new species of

enteropneust, Harrimania planktophilus (Hemichordata: Harrimaniidae) from barkley

sound. Biological Bulletin 202:182–191.

Cameron, C. B., J. R. Garey, and B. J. Swalla. 2000. Evolution of the chordate body plan: New

insights from phylogenetic analyses of deuterostome phyla. Proceedings of the National

Academy of Sciences of the United States of America 97:4469–4474.

Caron, J.-B., and D. A. Jackson. 2006. Taphonomy of the Greater Phyllopod Bed Community,

Burgess Shale. Palaios 21:451–465.

Caron, J.-B., and D. A. Jackson. 2008. Paleoecology of the Greater Phyllopod Bed community,

Burgess Shale. Palaeogeography, Palaeoclimatology, Palaeoecology 258:222–256.

263

Caron, J.-B., S. Conway Morris, and D. Shu. 2010. Tentaculate fossils from the Cambrian of

Canada (British Columbia) and China (Yunnan) interpreted as primitive deuterostomes.

PloS one 5:e9586.

Caron, J.-B., S. C. Morris, and C. B. Cameron. 2013. Tubicolous enteropneusts from the

Cambrian period. Nature 495:503–506.

Caron, J.-B., R. R. Gaines, C. Aria, M. G. Mángano, and M. Streng. 2014. A new phyllopod bed-

like assemblage from the Burgess Shale of the Canadian Rockies. Nature communications

5:3210.

Connell, J. H. 1961a. The influence of interspecific competition and other factors on the

distribution of the barnacle Chthamalus stellatus. Ecology 42:710–723.

———. 1961b. Effects of competition, predation by Thais lapillus, and other factors on natural

populations of the barnacle Balanus balanoides. Ecological Monographs 31:61–104.

Conway Morris, S. 1986. The community structure of the Middle Cambrian phyllopod bed

(Burgess Shale). Palaeontology 29:423–467.

Davidson, E. H., and D. H. Erwin. 2006. Gene regulatory networks and the evolution of animal

body plans. Science 311:796–800.

Erwin, D. H., M. Laflamme, S. M. Tweedt, E. A. Sperling, D. Pisani, and K. J. Peterson. 2011.

The Cambrian conundrum: early divergence and later ecological success in the early history

of animals. Science 334:1091–1097.

Gillis, J. A., J. H. Fritzenwanker, and C. J. Lowe. 2012. A stem-deuterostome origin of the

vertebrate pharyngeal transcriptional network. :237–246.

Gonzalez, P., K. R. Uhlinger, and C. J. Lowe. 2017. The adult body plan of indirect developing

hemichordates develops by adding a Hox-Patterned trunk to an anterior larval territory.

264

Current Biology 27:87–95.

Gould, S. J. 1989. Wonderful Life. W.W. Norton & Co., New York, N.Y., p.

Haug, J. T., D. E. Briggs, and C. Haug. 2012. Morphology and function in the Cambrian Burgess

Shale megacheiran arthropod Leanchoilia superlata and the application of a descriptive

matrix. BMC Evolutionary Biology 12.

Hu, S. X., M. Y. Zhu, M. Steiner, H. L. Luo, F. C. Zhao, and Q. Liu. 2010. Biodiversity and

taphonomy of the Early Cambrian Guanshan biota, eastern Yunnan. Science China Earth

Sciences 53:1765–1773.

Hubbell, S. P. 2001. The Unified Neutral Theory of Biodiversity and Biogeography.

Jensen, S., and G. E. Budd. 2000. A critical reappraisal of the fossil record of the bilaterian

phyla. Biological Reviews 75:253–295.

Komai, T. 1949. Internal structure of the pterobranch Atubaria heterolopha Sato, with an

appendix on the homology of the 'Notochord'. Proceedings of the Japan Academy 25:19–24.

Lefebvre, B., R. Lerosey-Aubril, T. Servais, and P. Van Roy. 2016. The Fezouata Biota: An

exceptional window on the Cambro-Ordovician faunal transition. Palaeogeography,

Palaeoclimatology, Palaeoecology 460:1–6.

Lowe, C. J., D. N. Clarke, D. M. Medeiros, D. S. Rokhsar, and J. Gerhart. 2015. The

deuterostome context of chordate origins. Nature 520:456–465.

Martin, E. L. O., B. Pittet, J. C. Gutiérrez-Marco, J. Vannier, K. El Hariri, R. Lerosey-Aubril, M.

Masrour, H. Nowak, T. Servais, T. R. A. Vandenbroucke, P. Van Roy, R. Vaucher, and B.

Lefebvre. 2016. The Lower Ordovician Fezouata Konservat-Lagerstätte from Morocco:

Age, environment and evolutionary perspectives. Gondwana Research 34:274–283.

Mitchell, C. E., M. J. Melchin, C. B. Cameron, and J. Maletz. 2013. Phylogenetic analysis

265

reveals that Rhabdopleura is an extant graptolite. Lethaia 46:34–56.

Morris, S. C. 1979. Middle Cambrian Polychaetes from the Burgess Shale of British Columbia.

Philosophical Transactions of the Royal Society B: Biological Sciences 285:227–274.

Murdock, D. J., S. E. Gabbott, G. Mayer, and M. a Purnell. 2014. Decay of velvet worms

(Onychophora), and bias in the fossil record of lobopodians. BMC Evolutionary Biology

14:1–9.

Nanglu, K., and J.-B. Caron. 2018. A new Burgess Shale polychaete and the origin of the annelid

head revisited. Current Biology 28:319–326.

Nanglu, K., J.-B. Caron, S. Conway Morris, and C. B. Cameron. 2016. Cambrian suspension-

feeding tubicolous hemichordates. BMC Biology 14.

O’Brien, L. J., and J. B. Caron. 2012. A new stalked Filter-Feeder from the Middle Cambrian

Burgess Shale, British Columbia, Canada. PLoS ONE 7.

O’Brien, L. J., and J. Caron. 2015. Paleocommunity analysis of the Burgess Shale Tulip Beds ,

Mount Stephen, British Columbia: Comparison with the Walcott Quarry and implications

for community variation in the Burgess Shale. Paleobiology:1–27.

Paine, R. T. 1974: Intertidal community structure. Experimental studies on the relationship

between an dominant competitor and its principal predator. Oecologia 15:93–120.

Peter, I. S., and E. H. Davidson. 2011. Evolution of gene regulatory networks controlling body

plan development. Cell 144:970–985.

Peterson, K. J., and D. J. Eernisse. 2016. The phylogeny, evolutionary developmental biology,

and paleobiology of the Deuterostomia: 25 years of new techniques , new discoveries , and

new ideas. Organisms Diversity & Evolution:401–418.

Peterson, K. J., M. R. Dietrich, and M. A. McPeek. 2009. MicroRNAs and metazoan

266

macroevolution: Insights into canalization, complexity, and the Cambrian explosion.

BioEssays 31:736–747.

Röttinger, E., and C. J. Lowe. 2012. Evolutionary crossroads in developmental biology :

hemichordates. Development 2475:2463–2475.

Van Roy, P., D. E. G. Briggs, and R. R. Gaines. 2015a. The Fezouata fossils of Morocco: an

extraordinary record of marine life in the Early Ordovician. Journal of the Geological

Society 172:541–549.

Van Roy, P., A. C. Daley, and D. E. G. Briggs. 2015b. Anomalocaridid trunk limb homology

revealed by a giant filter-feeder with paired flaps. Nature 522:77–80.

Ruppert, E. E. 2005. Key characters uniting hemichordates and chordates: homologies or

homoplasies? Canadian Journal of Zoology 83:8–23.

Sansom, R. S., S. E. Gabbott, and M. A. Purnell. 2010. Non-random decay of chordate characters

causes bias in fossil interpretation. Nature 463:797–800.

Sansom, R. S., S. E. Gabbott, and M. A. Purnell. 2013. Atlas of vertebrate decay: A visual and

taphonomic guide to fossil interpretation. Palaeontology 56:457–474.

Shubin, N., C. Tabin, and S. Carroll. 2009. Deep homology and the origins of evolutionary

novelty. Nature 457:818–823.

Szaniawski, H. 2005. Cambrian chaetognaths recognized in Burgess Shale fossils. Acta

Palaeontologica Polonica 50:1–8.

Walcott, C. D. 1911. Cambrian geology and paleontology. Middle Cambrian annelids.

Smithsonian Miscellaneous Collections 57:107–144.

Yang, J., J. Ortega-Hernández, N. J. Butterfield, and X. G. Zhang. 2013. Specialized appendages

in fuxianhuiids and the head organization of early euarthropods. Nature 494:468–471.

267

Zhao, F., J.-B. Caron, D. J. Bottjer, S. Hu, Z. Yin, and M. Zhu. 2013. Diversity and species

abundance patterns of the early Cambrian (Series 2, Stage 3) Chengjiang Biota from China.

Paleobiology 40:50–69.

Zhu, M. Y., Y. L. Zhao, and J. Y. Chen. 2002. Revision of the Cambrian discoidal animals

Stellostomites eumorphus and Pararotadiscus guizhouensis from South China. Geobios

35:165–185.

268